ita-latn-10mb-after-ppt-Dp-100mb-ckpt500_seed3407/checkpoint-15000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.4886859865025803,
  "eval_steps": 3000,
  "global_step": 15000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 7.415488815307617,
      "epoch": 0.0004962286621675268,
      "grad_norm": 1.0390625,
      "learning_rate": 2e-06,
      "loss": 6.9797,
      "mean_token_accuracy": 0.08811815902590751,
      "num_tokens": 11448.0,
      "step": 5
    },
    {
      "entropy": 7.423927688598633,
      "epoch": 0.0009924573243350536,
      "grad_norm": 1.046875,
      "learning_rate": 4.5e-06,
      "loss": 6.9808,
      "mean_token_accuracy": 0.08794851526618004,
      "num_tokens": 22646.0,
      "step": 10
    },
    {
      "entropy": 7.419756555557251,
      "epoch": 0.0014886859865025804,
      "grad_norm": 1.0859375,
      "learning_rate": 7e-06,
      "loss": 6.9117,
      "mean_token_accuracy": 0.08956757485866547,
      "num_tokens": 34011.0,
      "step": 15
    },
    {
      "entropy": 7.414588403701782,
      "epoch": 0.001984914648670107,
      "grad_norm": 0.9296875,
      "learning_rate": 9.5e-06,
      "loss": 6.9364,
      "mean_token_accuracy": 0.09392186626791954,
      "num_tokens": 44566.0,
      "step": 20
    },
    {
      "entropy": 7.422558164596557,
      "epoch": 0.002481143310837634,
      "grad_norm": 0.94921875,
      "learning_rate": 1.2e-05,
      "loss": 7.0066,
      "mean_token_accuracy": 0.08173666819930077,
      "num_tokens": 56274.0,
      "step": 25
    },
    {
      "entropy": 7.404311037063598,
      "epoch": 0.0029773719730051607,
      "grad_norm": 0.95703125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 6.9198,
      "mean_token_accuracy": 0.09323365837335587,
      "num_tokens": 66584.0,
      "step": 30
    },
    {
      "entropy": 7.403407335281372,
      "epoch": 0.0034736006351726877,
      "grad_norm": 0.98828125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 6.9536,
      "mean_token_accuracy": 0.08776896521449089,
      "num_tokens": 76948.0,
      "step": 35
    },
    {
      "entropy": 7.424892377853394,
      "epoch": 0.003969829297340214,
      "grad_norm": 0.82421875,
      "learning_rate": 1.95e-05,
      "loss": 6.9625,
      "mean_token_accuracy": 0.09215026199817658,
      "num_tokens": 87253.0,
      "step": 40
    },
    {
      "entropy": 7.402211713790893,
      "epoch": 0.004466057959507741,
      "grad_norm": 0.95703125,
      "learning_rate": 2.2e-05,
      "loss": 6.9664,
      "mean_token_accuracy": 0.0926709569990635,
      "num_tokens": 98273.0,
      "step": 45
    },
    {
      "entropy": 7.390836954116821,
      "epoch": 0.004962286621675268,
      "grad_norm": 1.0390625,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 6.8805,
      "mean_token_accuracy": 0.09474824145436286,
      "num_tokens": 107788.0,
      "step": 50
    },
    {
      "entropy": 7.361630630493164,
      "epoch": 0.0054585152838427945,
      "grad_norm": 0.91015625,
      "learning_rate": 2.7e-05,
      "loss": 6.9007,
      "mean_token_accuracy": 0.08987915515899658,
      "num_tokens": 118632.0,
      "step": 55
    },
    {
      "entropy": 7.329354906082154,
      "epoch": 0.0059547439460103215,
      "grad_norm": 0.89453125,
      "learning_rate": 2.95e-05,
      "loss": 6.8369,
      "mean_token_accuracy": 0.09355762079358101,
      "num_tokens": 128649.0,
      "step": 60
    },
    {
      "entropy": 7.331360340118408,
      "epoch": 0.0064509726081778485,
      "grad_norm": 0.95703125,
      "learning_rate": 3.2e-05,
      "loss": 6.8725,
      "mean_token_accuracy": 0.09848466441035271,
      "num_tokens": 138958.0,
      "step": 65
    },
    {
      "entropy": 7.316051912307739,
      "epoch": 0.0069472012703453755,
      "grad_norm": 0.91796875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 6.875,
      "mean_token_accuracy": 0.09506641924381257,
      "num_tokens": 151445.0,
      "step": 70
    },
    {
      "entropy": 7.357440185546875,
      "epoch": 0.007443429932512902,
      "grad_norm": 0.8515625,
      "learning_rate": 3.7e-05,
      "loss": 6.9168,
      "mean_token_accuracy": 0.0950050063431263,
      "num_tokens": 162239.0,
      "step": 75
    },
    {
      "entropy": 7.389816427230835,
      "epoch": 0.007939658594680429,
      "grad_norm": 1.0703125,
      "learning_rate": 3.95e-05,
      "loss": 6.8606,
      "mean_token_accuracy": 0.08998646065592766,
      "num_tokens": 173261.0,
      "step": 80
    },
    {
      "entropy": 7.374487590789795,
      "epoch": 0.008435887256847956,
      "grad_norm": 0.7578125,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 6.8612,
      "mean_token_accuracy": 0.08823171928524971,
      "num_tokens": 183857.0,
      "step": 85
    },
    {
      "entropy": 7.406818819046021,
      "epoch": 0.008932115919015483,
      "grad_norm": 0.98046875,
      "learning_rate": 4.45e-05,
      "loss": 7.0573,
      "mean_token_accuracy": 0.08514732420444489,
      "num_tokens": 196635.0,
      "step": 90
    },
    {
      "entropy": 7.393658542633057,
      "epoch": 0.00942834458118301,
      "grad_norm": 0.890625,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.0168,
      "mean_token_accuracy": 0.09471790939569473,
      "num_tokens": 208700.0,
      "step": 95
    },
    {
      "entropy": 7.416672229766846,
      "epoch": 0.009924573243350537,
      "grad_norm": 0.9609375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 6.9349,
      "mean_token_accuracy": 0.08931025043129921,
      "num_tokens": 218915.0,
      "step": 100
    },
    {
      "entropy": 7.340483951568603,
      "epoch": 0.010420801905518062,
      "grad_norm": 0.99609375,
      "learning_rate": 5.2e-05,
      "loss": 6.8607,
      "mean_token_accuracy": 0.0964167907834053,
      "num_tokens": 230211.0,
      "step": 105
    },
    {
      "entropy": 7.323957920074463,
      "epoch": 0.010917030567685589,
      "grad_norm": 0.984375,
      "learning_rate": 5.45e-05,
      "loss": 6.8819,
      "mean_token_accuracy": 0.10094936266541481,
      "num_tokens": 241174.0,
      "step": 110
    },
    {
      "entropy": 7.3142293930053714,
      "epoch": 0.011413259229853116,
      "grad_norm": 0.9375,
      "learning_rate": 5.7e-05,
      "loss": 6.8898,
      "mean_token_accuracy": 0.09185363873839378,
      "num_tokens": 251708.0,
      "step": 115
    },
    {
      "entropy": 7.415872716903687,
      "epoch": 0.011909487892020643,
      "grad_norm": 0.84765625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.0541,
      "mean_token_accuracy": 0.08325114101171494,
      "num_tokens": 263287.0,
      "step": 120
    },
    {
      "entropy": 7.43096809387207,
      "epoch": 0.01240571655418817,
      "grad_norm": 0.828125,
      "learning_rate": 6.2e-05,
      "loss": 6.9198,
      "mean_token_accuracy": 0.0861751176416874,
      "num_tokens": 275179.0,
      "step": 125
    },
    {
      "entropy": 7.31543345451355,
      "epoch": 0.012901945216355697,
      "grad_norm": 0.87890625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 6.7886,
      "mean_token_accuracy": 0.09708081930875778,
      "num_tokens": 287223.0,
      "step": 130
    },
    {
      "entropy": 7.303066396713257,
      "epoch": 0.013398173878523224,
      "grad_norm": 1.296875,
      "learning_rate": 6.7e-05,
      "loss": 6.8086,
      "mean_token_accuracy": 0.09785129725933076,
      "num_tokens": 297868.0,
      "step": 135
    },
    {
      "entropy": 7.2979652881622314,
      "epoch": 0.013894402540690751,
      "grad_norm": 0.78515625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 6.8927,
      "mean_token_accuracy": 0.09260747209191322,
      "num_tokens": 309702.0,
      "step": 140
    },
    {
      "entropy": 7.4375659942626955,
      "epoch": 0.014390631202858276,
      "grad_norm": 0.89453125,
      "learning_rate": 7.2e-05,
      "loss": 6.906,
      "mean_token_accuracy": 0.09100511819124221,
      "num_tokens": 321613.0,
      "step": 145
    },
    {
      "entropy": 7.383455657958985,
      "epoch": 0.014886859865025803,
      "grad_norm": 0.921875,
      "learning_rate": 7.45e-05,
      "loss": 6.8772,
      "mean_token_accuracy": 0.09483638480305671,
      "num_tokens": 332668.0,
      "step": 150
    },
    {
      "entropy": 7.347703647613526,
      "epoch": 0.01538308852719333,
      "grad_norm": 0.9609375,
      "learning_rate": 7.7e-05,
      "loss": 6.9471,
      "mean_token_accuracy": 0.09541060924530029,
      "num_tokens": 342579.0,
      "step": 155
    },
    {
      "entropy": 7.395112085342407,
      "epoch": 0.015879317189360857,
      "grad_norm": 1.0546875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 6.7956,
      "mean_token_accuracy": 0.09388362988829613,
      "num_tokens": 352271.0,
      "step": 160
    },
    {
      "entropy": 7.331162643432617,
      "epoch": 0.016375545851528384,
      "grad_norm": 0.9140625,
      "learning_rate": 8.2e-05,
      "loss": 6.932,
      "mean_token_accuracy": 0.09907099306583404,
      "num_tokens": 364033.0,
      "step": 165
    },
    {
      "entropy": 7.286223697662353,
      "epoch": 0.01687177451369591,
      "grad_norm": 0.8671875,
      "learning_rate": 8.450000000000001e-05,
      "loss": 6.8812,
      "mean_token_accuracy": 0.09280342161655426,
      "num_tokens": 373788.0,
      "step": 170
    },
    {
      "entropy": 7.398203802108765,
      "epoch": 0.01736800317586344,
      "grad_norm": 0.98046875,
      "learning_rate": 8.7e-05,
      "loss": 6.8526,
      "mean_token_accuracy": 0.09731495082378387,
      "num_tokens": 384883.0,
      "step": 175
    },
    {
      "entropy": 7.3108381748199465,
      "epoch": 0.017864231838030965,
      "grad_norm": 1.046875,
      "learning_rate": 8.95e-05,
      "loss": 6.7825,
      "mean_token_accuracy": 0.09594601169228553,
      "num_tokens": 396044.0,
      "step": 180
    },
    {
      "entropy": 7.364286279678344,
      "epoch": 0.018360460500198492,
      "grad_norm": 1.125,
      "learning_rate": 9.2e-05,
      "loss": 6.9289,
      "mean_token_accuracy": 0.09398710206151009,
      "num_tokens": 406995.0,
      "step": 185
    },
    {
      "entropy": 7.368616962432862,
      "epoch": 0.01885668916236602,
      "grad_norm": 1.03125,
      "learning_rate": 9.45e-05,
      "loss": 6.8242,
      "mean_token_accuracy": 0.09675540551543235,
      "num_tokens": 417238.0,
      "step": 190
    },
    {
      "entropy": 7.337496566772461,
      "epoch": 0.019352917824533546,
      "grad_norm": 0.9609375,
      "learning_rate": 9.7e-05,
      "loss": 6.869,
      "mean_token_accuracy": 0.09491769820451737,
      "num_tokens": 427263.0,
      "step": 195
    },
    {
      "entropy": 7.282687520980835,
      "epoch": 0.019849146486701073,
      "grad_norm": 0.8515625,
      "learning_rate": 9.95e-05,
      "loss": 6.7887,
      "mean_token_accuracy": 0.1014923520386219,
      "num_tokens": 438357.0,
      "step": 200
    },
    {
      "entropy": 7.343814468383789,
      "epoch": 0.0203453751488686,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000102,
      "loss": 6.8331,
      "mean_token_accuracy": 0.09941279143095016,
      "num_tokens": 448738.0,
      "step": 205
    },
    {
      "entropy": 7.334179782867432,
      "epoch": 0.020841603811036124,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00010449999999999999,
      "loss": 6.8794,
      "mean_token_accuracy": 0.09967483356595039,
      "num_tokens": 457873.0,
      "step": 210
    },
    {
      "entropy": 7.292327451705932,
      "epoch": 0.02133783247320365,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000107,
      "loss": 6.8026,
      "mean_token_accuracy": 0.09620003700256348,
      "num_tokens": 468640.0,
      "step": 215
    },
    {
      "entropy": 7.303198051452637,
      "epoch": 0.021834061135371178,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0001095,
      "loss": 6.8446,
      "mean_token_accuracy": 0.09441228732466697,
      "num_tokens": 479922.0,
      "step": 220
    },
    {
      "entropy": 7.285147094726563,
      "epoch": 0.022330289797538705,
      "grad_norm": 0.984375,
      "learning_rate": 0.000112,
      "loss": 6.7865,
      "mean_token_accuracy": 0.098848607391119,
      "num_tokens": 491056.0,
      "step": 225
    },
    {
      "entropy": 7.386207151412964,
      "epoch": 0.022826518459706232,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001145,
      "loss": 6.9114,
      "mean_token_accuracy": 0.09412190616130829,
      "num_tokens": 504365.0,
      "step": 230
    },
    {
      "entropy": 7.306986141204834,
      "epoch": 0.02332274712187376,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00011700000000000001,
      "loss": 6.7885,
      "mean_token_accuracy": 0.09399791210889816,
      "num_tokens": 514347.0,
      "step": 235
    },
    {
      "entropy": 7.284820985794068,
      "epoch": 0.023818975784041286,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00011949999999999999,
      "loss": 6.8463,
      "mean_token_accuracy": 0.09885015189647675,
      "num_tokens": 525912.0,
      "step": 240
    },
    {
      "entropy": 7.264670991897583,
      "epoch": 0.024315204446208813,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000122,
      "loss": 6.722,
      "mean_token_accuracy": 0.09802954122424126,
      "num_tokens": 535566.0,
      "step": 245
    },
    {
      "entropy": 7.293220567703247,
      "epoch": 0.02481143310837634,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0001245,
      "loss": 6.7569,
      "mean_token_accuracy": 0.09652359411120415,
      "num_tokens": 546474.0,
      "step": 250
    },
    {
      "entropy": 7.343952226638794,
      "epoch": 0.025307661770543867,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000127,
      "loss": 6.9085,
      "mean_token_accuracy": 0.09467220306396484,
      "num_tokens": 558132.0,
      "step": 255
    },
    {
      "entropy": 7.391662740707398,
      "epoch": 0.025803890432711394,
      "grad_norm": 1.25,
      "learning_rate": 0.0001295,
      "loss": 6.8717,
      "mean_token_accuracy": 0.08968425244092941,
      "num_tokens": 568309.0,
      "step": 260
    },
    {
      "entropy": 7.242248058319092,
      "epoch": 0.02630011909487892,
      "grad_norm": 1.109375,
      "learning_rate": 0.000132,
      "loss": 6.7985,
      "mean_token_accuracy": 0.09653588980436326,
      "num_tokens": 579701.0,
      "step": 265
    },
    {
      "entropy": 7.3309746265411375,
      "epoch": 0.026796347757046448,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00013450000000000002,
      "loss": 6.7519,
      "mean_token_accuracy": 0.09935276582837105,
      "num_tokens": 590832.0,
      "step": 270
    },
    {
      "entropy": 7.18915901184082,
      "epoch": 0.027292576419213975,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00013700000000000002,
      "loss": 6.7155,
      "mean_token_accuracy": 0.09612485915422439,
      "num_tokens": 601094.0,
      "step": 275
    },
    {
      "entropy": 7.255670118331909,
      "epoch": 0.027788805081381502,
      "grad_norm": 1.09375,
      "learning_rate": 0.0001395,
      "loss": 6.7341,
      "mean_token_accuracy": 0.1019403874874115,
      "num_tokens": 612027.0,
      "step": 280
    },
    {
      "entropy": 7.241375875473023,
      "epoch": 0.02828503374354903,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00014199999999999998,
      "loss": 6.7852,
      "mean_token_accuracy": 0.0944296196103096,
      "num_tokens": 623698.0,
      "step": 285
    },
    {
      "entropy": 7.326890134811402,
      "epoch": 0.028781262405716553,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0001445,
      "loss": 6.8125,
      "mean_token_accuracy": 0.09790188670158387,
      "num_tokens": 635631.0,
      "step": 290
    },
    {
      "entropy": 7.2972441673278805,
      "epoch": 0.02927749106788408,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000147,
      "loss": 6.8098,
      "mean_token_accuracy": 0.09671351835131645,
      "num_tokens": 647663.0,
      "step": 295
    },
    {
      "entropy": 7.226529216766357,
      "epoch": 0.029773719730051607,
      "grad_norm": 1.0,
      "learning_rate": 0.0001495,
      "loss": 6.7771,
      "mean_token_accuracy": 0.0953554093837738,
      "num_tokens": 659352.0,
      "step": 300
    },
    {
      "entropy": 7.249944972991943,
      "epoch": 0.030269948392219134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000152,
      "loss": 6.7277,
      "mean_token_accuracy": 0.10546225905418397,
      "num_tokens": 669379.0,
      "step": 305
    },
    {
      "entropy": 7.287822151184082,
      "epoch": 0.03076617705438666,
      "grad_norm": 1.171875,
      "learning_rate": 0.00015450000000000001,
      "loss": 6.7385,
      "mean_token_accuracy": 0.10098289772868156,
      "num_tokens": 680144.0,
      "step": 310
    },
    {
      "entropy": 7.2293524742126465,
      "epoch": 0.03126240571655419,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000157,
      "loss": 6.6844,
      "mean_token_accuracy": 0.09755392372608185,
      "num_tokens": 691165.0,
      "step": 315
    },
    {
      "entropy": 7.23696551322937,
      "epoch": 0.031758634378721715,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001595,
      "loss": 6.8136,
      "mean_token_accuracy": 0.09727144762873649,
      "num_tokens": 703767.0,
      "step": 320
    },
    {
      "entropy": 7.2687766551971436,
      "epoch": 0.032254863040889245,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000162,
      "loss": 6.7885,
      "mean_token_accuracy": 0.10152007564902306,
      "num_tokens": 714298.0,
      "step": 325
    },
    {
      "entropy": 7.237318611145019,
      "epoch": 0.03275109170305677,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00016450000000000001,
      "loss": 6.7841,
      "mean_token_accuracy": 0.10548736825585366,
      "num_tokens": 724135.0,
      "step": 330
    },
    {
      "entropy": 7.179241037368774,
      "epoch": 0.03324732036522429,
      "grad_norm": 0.9375,
      "learning_rate": 0.00016700000000000002,
      "loss": 6.8555,
      "mean_token_accuracy": 0.09413279443979264,
      "num_tokens": 735050.0,
      "step": 335
    },
    {
      "entropy": 7.315862798690796,
      "epoch": 0.03374354902739182,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00016950000000000003,
      "loss": 6.7585,
      "mean_token_accuracy": 0.09567313492298127,
      "num_tokens": 746073.0,
      "step": 340
    },
    {
      "entropy": 7.213669204711914,
      "epoch": 0.034239777689559346,
      "grad_norm": 1.015625,
      "learning_rate": 0.00017199999999999998,
      "loss": 6.7686,
      "mean_token_accuracy": 0.09640354067087173,
      "num_tokens": 758998.0,
      "step": 345
    },
    {
      "entropy": 7.203896570205688,
      "epoch": 0.03473600635172688,
      "grad_norm": 1.078125,
      "learning_rate": 0.00017449999999999999,
      "loss": 6.6867,
      "mean_token_accuracy": 0.09490930512547494,
      "num_tokens": 771608.0,
      "step": 350
    },
    {
      "entropy": 7.220716094970703,
      "epoch": 0.0352322350138944,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000177,
      "loss": 6.7711,
      "mean_token_accuracy": 0.09604462832212449,
      "num_tokens": 782394.0,
      "step": 355
    },
    {
      "entropy": 7.213645648956299,
      "epoch": 0.03572846367606193,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0001795,
      "loss": 6.6827,
      "mean_token_accuracy": 0.10222365930676461,
      "num_tokens": 792389.0,
      "step": 360
    },
    {
      "entropy": 7.195154285430908,
      "epoch": 0.036224692338229454,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000182,
      "loss": 6.7383,
      "mean_token_accuracy": 0.09877654984593391,
      "num_tokens": 803219.0,
      "step": 365
    },
    {
      "entropy": 7.26901273727417,
      "epoch": 0.036720921000396985,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001845,
      "loss": 6.8117,
      "mean_token_accuracy": 0.09464164227247238,
      "num_tokens": 814249.0,
      "step": 370
    },
    {
      "entropy": 7.248525381088257,
      "epoch": 0.03721714966256451,
      "grad_norm": 0.984375,
      "learning_rate": 0.000187,
      "loss": 6.7002,
      "mean_token_accuracy": 0.10740549117326736,
      "num_tokens": 825138.0,
      "step": 375
    },
    {
      "entropy": 7.151234674453735,
      "epoch": 0.03771337832473204,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001895,
      "loss": 6.7529,
      "mean_token_accuracy": 0.10170127227902412,
      "num_tokens": 836024.0,
      "step": 380
    },
    {
      "entropy": 7.221024084091186,
      "epoch": 0.03820960698689956,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000192,
      "loss": 6.7426,
      "mean_token_accuracy": 0.0970191240310669,
      "num_tokens": 847765.0,
      "step": 385
    },
    {
      "entropy": 7.190954208374023,
      "epoch": 0.03870583564906709,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001945,
      "loss": 6.7344,
      "mean_token_accuracy": 0.0974167302250862,
      "num_tokens": 859980.0,
      "step": 390
    },
    {
      "entropy": 7.299342060089112,
      "epoch": 0.039202064311234616,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00019700000000000002,
      "loss": 6.7494,
      "mean_token_accuracy": 0.09645001664757728,
      "num_tokens": 871514.0,
      "step": 395
    },
    {
      "entropy": 7.188382196426391,
      "epoch": 0.03969829297340215,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019950000000000002,
      "loss": 6.7508,
      "mean_token_accuracy": 0.09433658048510551,
      "num_tokens": 881984.0,
      "step": 400
    },
    {
      "entropy": 7.196045541763306,
      "epoch": 0.04019452163556967,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000202,
      "loss": 6.7063,
      "mean_token_accuracy": 0.10188237726688384,
      "num_tokens": 894198.0,
      "step": 405
    },
    {
      "entropy": 7.18197021484375,
      "epoch": 0.0406907502977372,
      "grad_norm": 1.0,
      "learning_rate": 0.00020449999999999998,
      "loss": 6.719,
      "mean_token_accuracy": 0.09976514428853989,
      "num_tokens": 905394.0,
      "step": 410
    },
    {
      "entropy": 7.248855495452881,
      "epoch": 0.041186978959904724,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000207,
      "loss": 6.7958,
      "mean_token_accuracy": 0.09452574029564857,
      "num_tokens": 917010.0,
      "step": 415
    },
    {
      "entropy": 7.155883693695069,
      "epoch": 0.04168320762207225,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002095,
      "loss": 6.6121,
      "mean_token_accuracy": 0.10424820482730865,
      "num_tokens": 926761.0,
      "step": 420
    },
    {
      "entropy": 7.060562229156494,
      "epoch": 0.04217943628423978,
      "grad_norm": 1.0625,
      "learning_rate": 0.000212,
      "loss": 6.6519,
      "mean_token_accuracy": 0.10032859295606614,
      "num_tokens": 938385.0,
      "step": 425
    },
    {
      "entropy": 7.134511184692383,
      "epoch": 0.0426756649464073,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002145,
      "loss": 6.6897,
      "mean_token_accuracy": 0.0993834190070629,
      "num_tokens": 949709.0,
      "step": 430
    },
    {
      "entropy": 7.11284236907959,
      "epoch": 0.04317189360857483,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00021700000000000002,
      "loss": 6.7173,
      "mean_token_accuracy": 0.09899322465062141,
      "num_tokens": 960837.0,
      "step": 435
    },
    {
      "entropy": 7.13794903755188,
      "epoch": 0.043668122270742356,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002195,
      "loss": 6.5545,
      "mean_token_accuracy": 0.10581014305353165,
      "num_tokens": 971036.0,
      "step": 440
    },
    {
      "entropy": 7.189764070510864,
      "epoch": 0.044164350932909886,
      "grad_norm": 1.078125,
      "learning_rate": 0.000222,
      "loss": 6.6988,
      "mean_token_accuracy": 0.09889093339443207,
      "num_tokens": 983057.0,
      "step": 445
    },
    {
      "entropy": 7.109373235702515,
      "epoch": 0.04466057959507741,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002245,
      "loss": 6.657,
      "mean_token_accuracy": 0.10294226482510567,
      "num_tokens": 994303.0,
      "step": 450
    },
    {
      "entropy": 7.106617975234985,
      "epoch": 0.04515680825724494,
      "grad_norm": 1.09375,
      "learning_rate": 0.00022700000000000002,
      "loss": 6.6683,
      "mean_token_accuracy": 0.1027058370411396,
      "num_tokens": 1005033.0,
      "step": 455
    },
    {
      "entropy": 7.166249561309814,
      "epoch": 0.045653036919412464,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00022950000000000002,
      "loss": 6.7313,
      "mean_token_accuracy": 0.10222981199622154,
      "num_tokens": 1015940.0,
      "step": 460
    },
    {
      "entropy": 7.1316859245300295,
      "epoch": 0.046149265581579994,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00023200000000000003,
      "loss": 6.6994,
      "mean_token_accuracy": 0.1031339094042778,
      "num_tokens": 1027582.0,
      "step": 465
    },
    {
      "entropy": 7.089751434326172,
      "epoch": 0.04664549424374752,
      "grad_norm": 0.9375,
      "learning_rate": 0.00023449999999999998,
      "loss": 6.6675,
      "mean_token_accuracy": 0.10097738355398178,
      "num_tokens": 1040046.0,
      "step": 470
    },
    {
      "entropy": 7.186102390289307,
      "epoch": 0.04714172290591505,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000237,
      "loss": 6.7444,
      "mean_token_accuracy": 0.09288590252399445,
      "num_tokens": 1050636.0,
      "step": 475
    },
    {
      "entropy": 7.142090320587158,
      "epoch": 0.04763795156808257,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002395,
      "loss": 6.6294,
      "mean_token_accuracy": 0.10634742677211761,
      "num_tokens": 1060748.0,
      "step": 480
    },
    {
      "entropy": 7.175977611541748,
      "epoch": 0.0481341802302501,
      "grad_norm": 0.875,
      "learning_rate": 0.000242,
      "loss": 6.6716,
      "mean_token_accuracy": 0.09640654549002647,
      "num_tokens": 1071940.0,
      "step": 485
    },
    {
      "entropy": 7.124125528335571,
      "epoch": 0.048630408892417626,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002445,
      "loss": 6.701,
      "mean_token_accuracy": 0.10127488672733306,
      "num_tokens": 1082962.0,
      "step": 490
    },
    {
      "entropy": 7.0944671630859375,
      "epoch": 0.04912663755458515,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000247,
      "loss": 6.6792,
      "mean_token_accuracy": 0.1065860316157341,
      "num_tokens": 1093082.0,
      "step": 495
    },
    {
      "entropy": 7.197715377807617,
      "epoch": 0.04962286621675268,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002495,
      "loss": 6.7682,
      "mean_token_accuracy": 0.10142254531383514,
      "num_tokens": 1103528.0,
      "step": 500
    },
    {
      "entropy": 7.073600006103516,
      "epoch": 0.0501190948789202,
      "grad_norm": 1.15625,
      "learning_rate": 0.000252,
      "loss": 6.6416,
      "mean_token_accuracy": 0.0960576593875885,
      "num_tokens": 1114813.0,
      "step": 505
    },
    {
      "entropy": 7.003829717636108,
      "epoch": 0.050615323541087734,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002545,
      "loss": 6.6403,
      "mean_token_accuracy": 0.09825952127575874,
      "num_tokens": 1125345.0,
      "step": 510
    },
    {
      "entropy": 7.108769464492798,
      "epoch": 0.05111155220325526,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000257,
      "loss": 6.6572,
      "mean_token_accuracy": 0.09955623000860214,
      "num_tokens": 1134985.0,
      "step": 515
    },
    {
      "entropy": 7.135790109634399,
      "epoch": 0.05160778086542279,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002595,
      "loss": 6.6783,
      "mean_token_accuracy": 0.10150780752301217,
      "num_tokens": 1145796.0,
      "step": 520
    },
    {
      "entropy": 7.187171220779419,
      "epoch": 0.05210400952759031,
      "grad_norm": 1.125,
      "learning_rate": 0.000262,
      "loss": 6.8131,
      "mean_token_accuracy": 0.09044179245829582,
      "num_tokens": 1156936.0,
      "step": 525
    },
    {
      "entropy": 7.090089416503906,
      "epoch": 0.05260023818975784,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026450000000000003,
      "loss": 6.6959,
      "mean_token_accuracy": 0.10054284557700158,
      "num_tokens": 1167526.0,
      "step": 530
    },
    {
      "entropy": 7.1854321479797365,
      "epoch": 0.053096466851925365,
      "grad_norm": 1.25,
      "learning_rate": 0.00026700000000000004,
      "loss": 6.7306,
      "mean_token_accuracy": 0.09534744620323181,
      "num_tokens": 1177647.0,
      "step": 535
    },
    {
      "entropy": 7.077204322814941,
      "epoch": 0.053592695514092896,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026950000000000005,
      "loss": 6.6923,
      "mean_token_accuracy": 0.1065493330359459,
      "num_tokens": 1188255.0,
      "step": 540
    },
    {
      "entropy": 7.047231197357178,
      "epoch": 0.05408892417626042,
      "grad_norm": 0.984375,
      "learning_rate": 0.00027200000000000005,
      "loss": 6.6221,
      "mean_token_accuracy": 0.11043694391846656,
      "num_tokens": 1199433.0,
      "step": 545
    },
    {
      "entropy": 7.008746385574341,
      "epoch": 0.05458515283842795,
      "grad_norm": 0.890625,
      "learning_rate": 0.0002745,
      "loss": 6.7064,
      "mean_token_accuracy": 0.09511427283287048,
      "num_tokens": 1210067.0,
      "step": 550
    },
    {
      "entropy": 7.019316959381103,
      "epoch": 0.055081381500595473,
      "grad_norm": 1.015625,
      "learning_rate": 0.000277,
      "loss": 6.5836,
      "mean_token_accuracy": 0.1026071272790432,
      "num_tokens": 1220639.0,
      "step": 555
    },
    {
      "entropy": 7.124681997299194,
      "epoch": 0.055577610162763004,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002795,
      "loss": 6.6339,
      "mean_token_accuracy": 0.10223110392689705,
      "num_tokens": 1233158.0,
      "step": 560
    },
    {
      "entropy": 7.022218942642212,
      "epoch": 0.05607383882493053,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00028199999999999997,
      "loss": 6.6128,
      "mean_token_accuracy": 0.10005637407302856,
      "num_tokens": 1244552.0,
      "step": 565
    },
    {
      "entropy": 7.1609320640563965,
      "epoch": 0.05657006748709806,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0002845,
      "loss": 6.7327,
      "mean_token_accuracy": 0.09698689877986907,
      "num_tokens": 1255478.0,
      "step": 570
    },
    {
      "entropy": 6.899557256698609,
      "epoch": 0.05706629614926558,
      "grad_norm": 1.015625,
      "learning_rate": 0.000287,
      "loss": 6.5604,
      "mean_token_accuracy": 0.10638210996985435,
      "num_tokens": 1267507.0,
      "step": 575
    },
    {
      "entropy": 7.014608669281006,
      "epoch": 0.057562524811433105,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002895,
      "loss": 6.5649,
      "mean_token_accuracy": 0.09933194518089294,
      "num_tokens": 1279377.0,
      "step": 580
    },
    {
      "entropy": 7.112942886352539,
      "epoch": 0.058058753473600636,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000292,
      "loss": 6.6557,
      "mean_token_accuracy": 0.10565379559993744,
      "num_tokens": 1290671.0,
      "step": 585
    },
    {
      "entropy": 6.903086757659912,
      "epoch": 0.05855498213576816,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0002945,
      "loss": 6.6323,
      "mean_token_accuracy": 0.10672209039330482,
      "num_tokens": 1302792.0,
      "step": 590
    },
    {
      "entropy": 7.101827001571655,
      "epoch": 0.05905121079793569,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000297,
      "loss": 6.5868,
      "mean_token_accuracy": 0.09854365140199661,
      "num_tokens": 1313315.0,
      "step": 595
    },
    {
      "entropy": 6.9200047016143795,
      "epoch": 0.05954743946010321,
      "grad_norm": 1.125,
      "learning_rate": 0.0002995,
      "loss": 6.6504,
      "mean_token_accuracy": 0.09922990500926972,
      "num_tokens": 1326358.0,
      "step": 600
    },
    {
      "entropy": 7.008511209487915,
      "epoch": 0.060043668122270744,
      "grad_norm": 1.0625,
      "learning_rate": 0.000302,
      "loss": 6.5085,
      "mean_token_accuracy": 0.1003392606973648,
      "num_tokens": 1336266.0,
      "step": 605
    },
    {
      "entropy": 6.907269525527954,
      "epoch": 0.06053989678443827,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003045,
      "loss": 6.4734,
      "mean_token_accuracy": 0.10566542968153954,
      "num_tokens": 1347610.0,
      "step": 610
    },
    {
      "entropy": 6.930313205718994,
      "epoch": 0.0610361254466058,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000307,
      "loss": 6.4398,
      "mean_token_accuracy": 0.10840422213077545,
      "num_tokens": 1358317.0,
      "step": 615
    },
    {
      "entropy": 6.901958322525024,
      "epoch": 0.06153235410877332,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003095,
      "loss": 6.5551,
      "mean_token_accuracy": 0.1008565068244934,
      "num_tokens": 1370126.0,
      "step": 620
    },
    {
      "entropy": 6.962115097045898,
      "epoch": 0.06202858277094085,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000312,
      "loss": 6.5148,
      "mean_token_accuracy": 0.10650759786367417,
      "num_tokens": 1381571.0,
      "step": 625
    },
    {
      "entropy": 6.91578860282898,
      "epoch": 0.06252481143310838,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003145,
      "loss": 6.5516,
      "mean_token_accuracy": 0.10437883734703064,
      "num_tokens": 1392486.0,
      "step": 630
    },
    {
      "entropy": 6.933821678161621,
      "epoch": 0.0630210400952759,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000317,
      "loss": 6.5571,
      "mean_token_accuracy": 0.10577636584639549,
      "num_tokens": 1403086.0,
      "step": 635
    },
    {
      "entropy": 6.9292055606842045,
      "epoch": 0.06351726875744343,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0003195,
      "loss": 6.5243,
      "mean_token_accuracy": 0.10535438284277916,
      "num_tokens": 1414537.0,
      "step": 640
    },
    {
      "entropy": 6.871757984161377,
      "epoch": 0.06401349741961096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000322,
      "loss": 6.5485,
      "mean_token_accuracy": 0.10128815025091172,
      "num_tokens": 1426998.0,
      "step": 645
    },
    {
      "entropy": 6.936267471313476,
      "epoch": 0.06450972608177849,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032450000000000003,
      "loss": 6.5753,
      "mean_token_accuracy": 0.10118880569934845,
      "num_tokens": 1437667.0,
      "step": 650
    },
    {
      "entropy": 6.886267423629761,
      "epoch": 0.065005954743946,
      "grad_norm": 1.0,
      "learning_rate": 0.00032700000000000003,
      "loss": 6.4528,
      "mean_token_accuracy": 0.10906932204961776,
      "num_tokens": 1448280.0,
      "step": 655
    },
    {
      "entropy": 6.921428108215332,
      "epoch": 0.06550218340611354,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032950000000000004,
      "loss": 6.5105,
      "mean_token_accuracy": 0.10349921584129333,
      "num_tokens": 1458608.0,
      "step": 660
    },
    {
      "entropy": 6.980496501922607,
      "epoch": 0.06599841206828107,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033200000000000005,
      "loss": 6.695,
      "mean_token_accuracy": 0.09470629766583442,
      "num_tokens": 1470435.0,
      "step": 665
    },
    {
      "entropy": 6.830832529067993,
      "epoch": 0.06649464073044858,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033450000000000005,
      "loss": 6.5469,
      "mean_token_accuracy": 0.10282312780618667,
      "num_tokens": 1481959.0,
      "step": 670
    },
    {
      "entropy": 6.974913597106934,
      "epoch": 0.06699086939261611,
      "grad_norm": 1.015625,
      "learning_rate": 0.000337,
      "loss": 6.5976,
      "mean_token_accuracy": 0.09841747283935547,
      "num_tokens": 1493798.0,
      "step": 675
    },
    {
      "entropy": 6.887060308456421,
      "epoch": 0.06748709805478365,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003395,
      "loss": 6.403,
      "mean_token_accuracy": 0.10493207052350044,
      "num_tokens": 1502969.0,
      "step": 680
    },
    {
      "entropy": 6.897101402282715,
      "epoch": 0.06798332671695118,
      "grad_norm": 1.1875,
      "learning_rate": 0.000342,
      "loss": 6.5595,
      "mean_token_accuracy": 0.10253233015537262,
      "num_tokens": 1513825.0,
      "step": 685
    },
    {
      "entropy": 6.822127819061279,
      "epoch": 0.06847955537911869,
      "grad_norm": 0.953125,
      "learning_rate": 0.00034449999999999997,
      "loss": 6.4575,
      "mean_token_accuracy": 0.10815273150801659,
      "num_tokens": 1525328.0,
      "step": 690
    },
    {
      "entropy": 6.891428852081299,
      "epoch": 0.06897578404128622,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000347,
      "loss": 6.4859,
      "mean_token_accuracy": 0.10564783737063407,
      "num_tokens": 1537383.0,
      "step": 695
    },
    {
      "entropy": 6.913006925582886,
      "epoch": 0.06947201270345375,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003495,
      "loss": 6.5826,
      "mean_token_accuracy": 0.10424128845334053,
      "num_tokens": 1549671.0,
      "step": 700
    },
    {
      "entropy": 6.7738439559936525,
      "epoch": 0.06996824136562128,
      "grad_norm": 1.171875,
      "learning_rate": 0.000352,
      "loss": 6.4562,
      "mean_token_accuracy": 0.1023649513721466,
      "num_tokens": 1559974.0,
      "step": 705
    },
    {
      "entropy": 6.948292827606201,
      "epoch": 0.0704644700277888,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003545,
      "loss": 6.5642,
      "mean_token_accuracy": 0.09899581745266914,
      "num_tokens": 1570890.0,
      "step": 710
    },
    {
      "entropy": 6.786388778686524,
      "epoch": 0.07096069868995633,
      "grad_norm": 1.15625,
      "learning_rate": 0.000357,
      "loss": 6.473,
      "mean_token_accuracy": 0.11216894015669823,
      "num_tokens": 1581434.0,
      "step": 715
    },
    {
      "entropy": 6.885098886489868,
      "epoch": 0.07145692735212386,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003595,
      "loss": 6.5411,
      "mean_token_accuracy": 0.1010577216744423,
      "num_tokens": 1593710.0,
      "step": 720
    },
    {
      "entropy": 6.914775133132935,
      "epoch": 0.07195315601429139,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000362,
      "loss": 6.5039,
      "mean_token_accuracy": 0.10822037607431412,
      "num_tokens": 1604635.0,
      "step": 725
    },
    {
      "entropy": 6.96195125579834,
      "epoch": 0.07244938467645891,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003645,
      "loss": 6.5621,
      "mean_token_accuracy": 0.10865710079669952,
      "num_tokens": 1615258.0,
      "step": 730
    },
    {
      "entropy": 6.818844604492187,
      "epoch": 0.07294561333862644,
      "grad_norm": 1.0,
      "learning_rate": 0.000367,
      "loss": 6.4657,
      "mean_token_accuracy": 0.1062205784022808,
      "num_tokens": 1625942.0,
      "step": 735
    },
    {
      "entropy": 6.773487854003906,
      "epoch": 0.07344184200079397,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003695,
      "loss": 6.4609,
      "mean_token_accuracy": 0.11281525269150734,
      "num_tokens": 1636270.0,
      "step": 740
    },
    {
      "entropy": 6.791012191772461,
      "epoch": 0.07393807066296149,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000372,
      "loss": 6.442,
      "mean_token_accuracy": 0.10923936069011689,
      "num_tokens": 1647029.0,
      "step": 745
    },
    {
      "entropy": 6.894232225418091,
      "epoch": 0.07443429932512902,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003745,
      "loss": 6.6043,
      "mean_token_accuracy": 0.10349742248654366,
      "num_tokens": 1659483.0,
      "step": 750
    },
    {
      "entropy": 6.763030767440796,
      "epoch": 0.07493052798729655,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000377,
      "loss": 6.4626,
      "mean_token_accuracy": 0.10359987020492553,
      "num_tokens": 1669425.0,
      "step": 755
    },
    {
      "entropy": 6.839314746856689,
      "epoch": 0.07542675664946408,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003795,
      "loss": 6.4669,
      "mean_token_accuracy": 0.10515316724777221,
      "num_tokens": 1680426.0,
      "step": 760
    },
    {
      "entropy": 6.7409745216369625,
      "epoch": 0.0759229853116316,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000382,
      "loss": 6.49,
      "mean_token_accuracy": 0.1022932469844818,
      "num_tokens": 1692803.0,
      "step": 765
    },
    {
      "entropy": 6.86904993057251,
      "epoch": 0.07641921397379912,
      "grad_norm": 0.90625,
      "learning_rate": 0.0003845,
      "loss": 6.5127,
      "mean_token_accuracy": 0.10286033898591995,
      "num_tokens": 1703919.0,
      "step": 770
    },
    {
      "entropy": 6.803786706924439,
      "epoch": 0.07691544263596665,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00038700000000000003,
      "loss": 6.4902,
      "mean_token_accuracy": 0.11186280027031899,
      "num_tokens": 1714487.0,
      "step": 775
    },
    {
      "entropy": 6.829148006439209,
      "epoch": 0.07741167129813419,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038950000000000003,
      "loss": 6.5375,
      "mean_token_accuracy": 0.103410904109478,
      "num_tokens": 1726601.0,
      "step": 780
    },
    {
      "entropy": 6.77627649307251,
      "epoch": 0.0779078999603017,
      "grad_norm": 0.96875,
      "learning_rate": 0.00039200000000000004,
      "loss": 6.4293,
      "mean_token_accuracy": 0.10090705007314682,
      "num_tokens": 1738743.0,
      "step": 785
    },
    {
      "entropy": 6.75617151260376,
      "epoch": 0.07840412862246923,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039450000000000005,
      "loss": 6.509,
      "mean_token_accuracy": 0.10762425512075424,
      "num_tokens": 1748286.0,
      "step": 790
    },
    {
      "entropy": 6.78504114151001,
      "epoch": 0.07890035728463676,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00039700000000000005,
      "loss": 6.5497,
      "mean_token_accuracy": 0.10279078707098961,
      "num_tokens": 1758917.0,
      "step": 795
    },
    {
      "entropy": 6.834147977828979,
      "epoch": 0.0793965859468043,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003995,
      "loss": 6.3768,
      "mean_token_accuracy": 0.11014361456036567,
      "num_tokens": 1769179.0,
      "step": 800
    },
    {
      "entropy": 6.686023187637329,
      "epoch": 0.07989281460897181,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000402,
      "loss": 6.5004,
      "mean_token_accuracy": 0.09990919530391693,
      "num_tokens": 1782369.0,
      "step": 805
    },
    {
      "entropy": 6.789707326889038,
      "epoch": 0.08038904327113934,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004045,
      "loss": 6.5087,
      "mean_token_accuracy": 0.11087186112999917,
      "num_tokens": 1795992.0,
      "step": 810
    },
    {
      "entropy": 6.767456674575806,
      "epoch": 0.08088527193330687,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00040699999999999997,
      "loss": 6.4063,
      "mean_token_accuracy": 0.11133067905902863,
      "num_tokens": 1806938.0,
      "step": 815
    },
    {
      "entropy": 6.6453076839447025,
      "epoch": 0.0813815005954744,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004095,
      "loss": 6.3335,
      "mean_token_accuracy": 0.1083066463470459,
      "num_tokens": 1817663.0,
      "step": 820
    },
    {
      "entropy": 6.804880285263062,
      "epoch": 0.08187772925764192,
      "grad_norm": 0.984375,
      "learning_rate": 0.000412,
      "loss": 6.4937,
      "mean_token_accuracy": 0.10573754906654358,
      "num_tokens": 1828254.0,
      "step": 825
    },
    {
      "entropy": 6.744491100311279,
      "epoch": 0.08237395791980945,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004145,
      "loss": 6.4121,
      "mean_token_accuracy": 0.10808726996183396,
      "num_tokens": 1839001.0,
      "step": 830
    },
    {
      "entropy": 6.780965948104859,
      "epoch": 0.08287018658197698,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000417,
      "loss": 6.5072,
      "mean_token_accuracy": 0.1064543254673481,
      "num_tokens": 1849906.0,
      "step": 835
    },
    {
      "entropy": 6.638266897201538,
      "epoch": 0.0833664152441445,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004195,
      "loss": 6.4146,
      "mean_token_accuracy": 0.10410640761256218,
      "num_tokens": 1860982.0,
      "step": 840
    },
    {
      "entropy": 6.7845916748046875,
      "epoch": 0.08386264390631203,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000422,
      "loss": 6.3668,
      "mean_token_accuracy": 0.10832294374704361,
      "num_tokens": 1872177.0,
      "step": 845
    },
    {
      "entropy": 6.6455484390258786,
      "epoch": 0.08435887256847956,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004245,
      "loss": 6.345,
      "mean_token_accuracy": 0.10717036500573159,
      "num_tokens": 1882783.0,
      "step": 850
    },
    {
      "entropy": 6.82792387008667,
      "epoch": 0.08485510123064709,
      "grad_norm": 1.09375,
      "learning_rate": 0.000427,
      "loss": 6.521,
      "mean_token_accuracy": 0.0995104432106018,
      "num_tokens": 1893706.0,
      "step": 855
    },
    {
      "entropy": 6.6573139190673825,
      "epoch": 0.0853513298928146,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004295,
      "loss": 6.399,
      "mean_token_accuracy": 0.10572489872574806,
      "num_tokens": 1903852.0,
      "step": 860
    },
    {
      "entropy": 6.720736408233643,
      "epoch": 0.08584755855498213,
      "grad_norm": 1.0625,
      "learning_rate": 0.000432,
      "loss": 6.4272,
      "mean_token_accuracy": 0.10979009121656418,
      "num_tokens": 1914157.0,
      "step": 865
    },
    {
      "entropy": 6.719715213775634,
      "epoch": 0.08634378721714966,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004345,
      "loss": 6.3646,
      "mean_token_accuracy": 0.10477298423647881,
      "num_tokens": 1924684.0,
      "step": 870
    },
    {
      "entropy": 6.765842390060425,
      "epoch": 0.0868400158793172,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000437,
      "loss": 6.5126,
      "mean_token_accuracy": 0.1022205650806427,
      "num_tokens": 1935657.0,
      "step": 875
    },
    {
      "entropy": 6.674496650695801,
      "epoch": 0.08733624454148471,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004395,
      "loss": 6.3866,
      "mean_token_accuracy": 0.10951956436038017,
      "num_tokens": 1946735.0,
      "step": 880
    },
    {
      "entropy": 6.771207475662232,
      "epoch": 0.08783247320365224,
      "grad_norm": 1.046875,
      "learning_rate": 0.000442,
      "loss": 6.558,
      "mean_token_accuracy": 0.10081006437540055,
      "num_tokens": 1957871.0,
      "step": 885
    },
    {
      "entropy": 6.6469238758087155,
      "epoch": 0.08832870186581977,
      "grad_norm": 1.125,
      "learning_rate": 0.0004445,
      "loss": 6.4001,
      "mean_token_accuracy": 0.10841071009635925,
      "num_tokens": 1969032.0,
      "step": 890
    },
    {
      "entropy": 6.773042964935303,
      "epoch": 0.0888249305279873,
      "grad_norm": 1.046875,
      "learning_rate": 0.000447,
      "loss": 6.4253,
      "mean_token_accuracy": 0.10482999682426453,
      "num_tokens": 1980212.0,
      "step": 895
    },
    {
      "entropy": 6.5782490253448485,
      "epoch": 0.08932115919015482,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00044950000000000003,
      "loss": 6.3457,
      "mean_token_accuracy": 0.1083353117108345,
      "num_tokens": 1990333.0,
      "step": 900
    },
    {
      "entropy": 6.723943090438842,
      "epoch": 0.08981738785232235,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045200000000000004,
      "loss": 6.4208,
      "mean_token_accuracy": 0.11527454107999802,
      "num_tokens": 2001486.0,
      "step": 905
    },
    {
      "entropy": 6.647228240966797,
      "epoch": 0.09031361651448988,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045450000000000004,
      "loss": 6.4152,
      "mean_token_accuracy": 0.10908102542161942,
      "num_tokens": 2012535.0,
      "step": 910
    },
    {
      "entropy": 6.65789384841919,
      "epoch": 0.0908098451766574,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045700000000000005,
      "loss": 6.3697,
      "mean_token_accuracy": 0.11192298531532288,
      "num_tokens": 2024116.0,
      "step": 915
    },
    {
      "entropy": 6.552388858795166,
      "epoch": 0.09130607383882493,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00045950000000000006,
      "loss": 6.4129,
      "mean_token_accuracy": 0.10573532730340958,
      "num_tokens": 2036341.0,
      "step": 920
    },
    {
      "entropy": 6.772451591491699,
      "epoch": 0.09180230250099246,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000462,
      "loss": 6.3831,
      "mean_token_accuracy": 0.11066723242402077,
      "num_tokens": 2046768.0,
      "step": 925
    },
    {
      "entropy": 6.601136875152588,
      "epoch": 0.09229853116315999,
      "grad_norm": 1.0,
      "learning_rate": 0.0004645,
      "loss": 6.3447,
      "mean_token_accuracy": 0.10959907621145248,
      "num_tokens": 2057315.0,
      "step": 930
    },
    {
      "entropy": 6.595295667648315,
      "epoch": 0.0927947598253275,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000467,
      "loss": 6.3258,
      "mean_token_accuracy": 0.1111043132841587,
      "num_tokens": 2068267.0,
      "step": 935
    },
    {
      "entropy": 6.6356977939605715,
      "epoch": 0.09329098848749504,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004695,
      "loss": 6.4018,
      "mean_token_accuracy": 0.10635230094194412,
      "num_tokens": 2078271.0,
      "step": 940
    },
    {
      "entropy": 6.570490837097168,
      "epoch": 0.09378721714966257,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000472,
      "loss": 6.3186,
      "mean_token_accuracy": 0.10399494022130966,
      "num_tokens": 2088381.0,
      "step": 945
    },
    {
      "entropy": 6.635789442062378,
      "epoch": 0.0942834458118301,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004745,
      "loss": 6.2807,
      "mean_token_accuracy": 0.10803479552268982,
      "num_tokens": 2098899.0,
      "step": 950
    },
    {
      "entropy": 6.580207061767578,
      "epoch": 0.09477967447399761,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000477,
      "loss": 6.4469,
      "mean_token_accuracy": 0.1090017631649971,
      "num_tokens": 2110057.0,
      "step": 955
    },
    {
      "entropy": 6.590322017669678,
      "epoch": 0.09527590313616514,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004795,
      "loss": 6.3534,
      "mean_token_accuracy": 0.10890609472990036,
      "num_tokens": 2121758.0,
      "step": 960
    },
    {
      "entropy": 6.626723718643189,
      "epoch": 0.09577213179833267,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000482,
      "loss": 6.3606,
      "mean_token_accuracy": 0.11137763261795045,
      "num_tokens": 2131840.0,
      "step": 965
    },
    {
      "entropy": 6.4795605659484865,
      "epoch": 0.0962683604605002,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004845,
      "loss": 6.3243,
      "mean_token_accuracy": 0.11041747704148293,
      "num_tokens": 2142084.0,
      "step": 970
    },
    {
      "entropy": 6.671319103240966,
      "epoch": 0.09676458912266772,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000487,
      "loss": 6.3937,
      "mean_token_accuracy": 0.10798183381557465,
      "num_tokens": 2152785.0,
      "step": 975
    },
    {
      "entropy": 6.661399173736572,
      "epoch": 0.09726081778483525,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004895,
      "loss": 6.3784,
      "mean_token_accuracy": 0.109221463650465,
      "num_tokens": 2163774.0,
      "step": 980
    },
    {
      "entropy": 6.505852460861206,
      "epoch": 0.09775704644700278,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000492,
      "loss": 6.2916,
      "mean_token_accuracy": 0.1088893674314022,
      "num_tokens": 2175378.0,
      "step": 985
    },
    {
      "entropy": 6.755297660827637,
      "epoch": 0.0982532751091703,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004945,
      "loss": 6.5117,
      "mean_token_accuracy": 0.1058343231678009,
      "num_tokens": 2186510.0,
      "step": 990
    },
    {
      "entropy": 6.522902965545654,
      "epoch": 0.09874950377133783,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497,
      "loss": 6.2971,
      "mean_token_accuracy": 0.11370693743228913,
      "num_tokens": 2197482.0,
      "step": 995
    },
    {
      "entropy": 6.5511187553405765,
      "epoch": 0.09924573243350536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995,
      "loss": 6.3049,
      "mean_token_accuracy": 0.10995083898305893,
      "num_tokens": 2208329.0,
      "step": 1000
    },
    {
      "entropy": 6.5022056102752686,
      "epoch": 0.09974196109567289,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999999982145551,
      "loss": 6.2196,
      "mean_token_accuracy": 0.10750824585556984,
      "num_tokens": 2218632.0,
      "step": 1005
    },
    {
      "entropy": 6.522266244888305,
      "epoch": 0.1002381897578404,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999999909611855,
      "loss": 6.3288,
      "mean_token_accuracy": 0.10453014001250267,
      "num_tokens": 2231335.0,
      "step": 1010
    },
    {
      "entropy": 6.613006448745727,
      "epoch": 0.10073441842000794,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999999781283008,
      "loss": 6.3162,
      "mean_token_accuracy": 0.11068271622061729,
      "num_tokens": 2241704.0,
      "step": 1015
    },
    {
      "entropy": 6.541505002975464,
      "epoch": 0.10123064708217547,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999999597159017,
      "loss": 6.3683,
      "mean_token_accuracy": 0.10505040511488914,
      "num_tokens": 2252310.0,
      "step": 1020
    },
    {
      "entropy": 6.5655745506286625,
      "epoch": 0.101726875744343,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999999357239882,
      "loss": 6.289,
      "mean_token_accuracy": 0.10687798410654067,
      "num_tokens": 2262270.0,
      "step": 1025
    },
    {
      "entropy": 6.599134349822998,
      "epoch": 0.10222310440651051,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999999061525612,
      "loss": 6.4258,
      "mean_token_accuracy": 0.10728801861405372,
      "num_tokens": 2273986.0,
      "step": 1030
    },
    {
      "entropy": 6.644435930252075,
      "epoch": 0.10271933306867805,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999998710016214,
      "loss": 6.3152,
      "mean_token_accuracy": 0.10951637625694274,
      "num_tokens": 2283430.0,
      "step": 1035
    },
    {
      "entropy": 6.572604465484619,
      "epoch": 0.10321556173084558,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999998302711696,
      "loss": 6.3347,
      "mean_token_accuracy": 0.10473796129226684,
      "num_tokens": 2294201.0,
      "step": 1040
    },
    {
      "entropy": 6.451688957214356,
      "epoch": 0.1037117903930131,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999997839612069,
      "loss": 6.2316,
      "mean_token_accuracy": 0.10967583358287811,
      "num_tokens": 2303876.0,
      "step": 1045
    },
    {
      "entropy": 6.53422532081604,
      "epoch": 0.10420801905518062,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999997320717342,
      "loss": 6.2748,
      "mean_token_accuracy": 0.11531499549746513,
      "num_tokens": 2313713.0,
      "step": 1050
    },
    {
      "entropy": 6.462176990509033,
      "epoch": 0.10470424771734815,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999996746027531,
      "loss": 6.281,
      "mean_token_accuracy": 0.1057841993868351,
      "num_tokens": 2324862.0,
      "step": 1055
    },
    {
      "entropy": 6.63836817741394,
      "epoch": 0.10520047637951568,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499999611554265,
      "loss": 6.342,
      "mean_token_accuracy": 0.11243951395153999,
      "num_tokens": 2334974.0,
      "step": 1060
    },
    {
      "entropy": 6.527207708358764,
      "epoch": 0.10569670504168321,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999995429262712,
      "loss": 6.3814,
      "mean_token_accuracy": 0.11008421406149864,
      "num_tokens": 2345945.0,
      "step": 1065
    },
    {
      "entropy": 6.573855066299439,
      "epoch": 0.10619293370385073,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999994687187737,
      "loss": 6.3162,
      "mean_token_accuracy": 0.10868648067116737,
      "num_tokens": 2356749.0,
      "step": 1070
    },
    {
      "entropy": 6.601923894882202,
      "epoch": 0.10668916236601826,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999993889317742,
      "loss": 6.3894,
      "mean_token_accuracy": 0.10912014171481133,
      "num_tokens": 2368181.0,
      "step": 1075
    },
    {
      "entropy": 6.519433784484863,
      "epoch": 0.10718539102818579,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999993035652746,
      "loss": 6.2988,
      "mean_token_accuracy": 0.10856023505330085,
      "num_tokens": 2377737.0,
      "step": 1080
    },
    {
      "entropy": 6.498434209823609,
      "epoch": 0.10768161969035331,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999992126192771,
      "loss": 6.2253,
      "mean_token_accuracy": 0.11343928501009941,
      "num_tokens": 2388944.0,
      "step": 1085
    },
    {
      "entropy": 6.506166744232178,
      "epoch": 0.10817784835252084,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499999116093784,
      "loss": 6.3021,
      "mean_token_accuracy": 0.11269065588712693,
      "num_tokens": 2400758.0,
      "step": 1090
    },
    {
      "entropy": 6.475096273422241,
      "epoch": 0.10867407701468837,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999990139887976,
      "loss": 6.2988,
      "mean_token_accuracy": 0.11071598455309868,
      "num_tokens": 2411916.0,
      "step": 1095
    },
    {
      "entropy": 6.6090761661529545,
      "epoch": 0.1091703056768559,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999989063043206,
      "loss": 6.3162,
      "mean_token_accuracy": 0.10899331420660019,
      "num_tokens": 2423310.0,
      "step": 1100
    },
    {
      "entropy": 6.4430852890014645,
      "epoch": 0.10966653433902342,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999987930403555,
      "loss": 6.255,
      "mean_token_accuracy": 0.11132511496543884,
      "num_tokens": 2432100.0,
      "step": 1105
    },
    {
      "entropy": 6.497279357910156,
      "epoch": 0.11016276300119095,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999986741969051,
      "loss": 6.3283,
      "mean_token_accuracy": 0.10885677859187126,
      "num_tokens": 2443563.0,
      "step": 1110
    },
    {
      "entropy": 6.568647575378418,
      "epoch": 0.11065899166335848,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999985497739724,
      "loss": 6.3328,
      "mean_token_accuracy": 0.10904523879289627,
      "num_tokens": 2455435.0,
      "step": 1115
    },
    {
      "entropy": 6.526462554931641,
      "epoch": 0.11115522032552601,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999984197715606,
      "loss": 6.2693,
      "mean_token_accuracy": 0.11513903066515922,
      "num_tokens": 2466173.0,
      "step": 1120
    },
    {
      "entropy": 6.406380701065063,
      "epoch": 0.11165144898769352,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999982841896728,
      "loss": 6.2217,
      "mean_token_accuracy": 0.11128725931048393,
      "num_tokens": 2476349.0,
      "step": 1125
    },
    {
      "entropy": 6.5459887981414795,
      "epoch": 0.11214767764986106,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999981430283123,
      "loss": 6.3545,
      "mean_token_accuracy": 0.11260213553905488,
      "num_tokens": 2487864.0,
      "step": 1130
    },
    {
      "entropy": 6.459234380722046,
      "epoch": 0.11264390631202859,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999979962874827,
      "loss": 6.2778,
      "mean_token_accuracy": 0.11224315464496612,
      "num_tokens": 2499188.0,
      "step": 1135
    },
    {
      "entropy": 6.541913795471191,
      "epoch": 0.11314013497419612,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999978439671877,
      "loss": 6.2679,
      "mean_token_accuracy": 0.11238265410065651,
      "num_tokens": 2509813.0,
      "step": 1140
    },
    {
      "entropy": 6.452944564819336,
      "epoch": 0.11363636363636363,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999976860674309,
      "loss": 6.2954,
      "mean_token_accuracy": 0.1079881377518177,
      "num_tokens": 2520813.0,
      "step": 1145
    },
    {
      "entropy": 6.451686000823974,
      "epoch": 0.11413259229853116,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999975225882164,
      "loss": 6.2857,
      "mean_token_accuracy": 0.11005399227142335,
      "num_tokens": 2531802.0,
      "step": 1150
    },
    {
      "entropy": 6.419873952865601,
      "epoch": 0.1146288209606987,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499997353529548,
      "loss": 6.0989,
      "mean_token_accuracy": 0.12534342482686042,
      "num_tokens": 2541935.0,
      "step": 1155
    },
    {
      "entropy": 6.41290922164917,
      "epoch": 0.11512504962286621,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999971788914303,
      "loss": 6.209,
      "mean_token_accuracy": 0.11427329108119011,
      "num_tokens": 2553224.0,
      "step": 1160
    },
    {
      "entropy": 6.4005992889404295,
      "epoch": 0.11562127828503374,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999969986738672,
      "loss": 6.2408,
      "mean_token_accuracy": 0.1071538008749485,
      "num_tokens": 2565355.0,
      "step": 1165
    },
    {
      "entropy": 6.517459058761597,
      "epoch": 0.11611750694720127,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999968128768634,
      "loss": 6.3245,
      "mean_token_accuracy": 0.1080057941377163,
      "num_tokens": 2575962.0,
      "step": 1170
    },
    {
      "entropy": 6.512309551239014,
      "epoch": 0.1166137356093688,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999966215004236,
      "loss": 6.3356,
      "mean_token_accuracy": 0.11455044671893119,
      "num_tokens": 2588136.0,
      "step": 1175
    },
    {
      "entropy": 6.473892450332642,
      "epoch": 0.11710996427153632,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999964245445522,
      "loss": 6.2264,
      "mean_token_accuracy": 0.11278848052024841,
      "num_tokens": 2597785.0,
      "step": 1180
    },
    {
      "entropy": 6.45322551727295,
      "epoch": 0.11760619293370385,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999962220092545,
      "loss": 6.2625,
      "mean_token_accuracy": 0.12006586417555809,
      "num_tokens": 2609738.0,
      "step": 1185
    },
    {
      "entropy": 6.329366683959961,
      "epoch": 0.11810242159587138,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999960138945351,
      "loss": 6.1518,
      "mean_token_accuracy": 0.12184964567422867,
      "num_tokens": 2620179.0,
      "step": 1190
    },
    {
      "entropy": 6.402722692489624,
      "epoch": 0.11859865025803891,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999958002003996,
      "loss": 6.1977,
      "mean_token_accuracy": 0.1172001026570797,
      "num_tokens": 2631550.0,
      "step": 1195
    },
    {
      "entropy": 6.43868670463562,
      "epoch": 0.11909487892020643,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999955809268529,
      "loss": 6.2334,
      "mean_token_accuracy": 0.1126227967441082,
      "num_tokens": 2642467.0,
      "step": 1200
    },
    {
      "entropy": 6.404077625274658,
      "epoch": 0.11959110758237396,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999953560739008,
      "loss": 6.2134,
      "mean_token_accuracy": 0.11190939992666245,
      "num_tokens": 2652132.0,
      "step": 1205
    },
    {
      "entropy": 6.447581481933594,
      "epoch": 0.12008733624454149,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999951256415485,
      "loss": 6.2707,
      "mean_token_accuracy": 0.11127389594912529,
      "num_tokens": 2663336.0,
      "step": 1210
    },
    {
      "entropy": 6.5261087894439695,
      "epoch": 0.12058356490670902,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499994889629802,
      "loss": 6.3448,
      "mean_token_accuracy": 0.10134708434343338,
      "num_tokens": 2674201.0,
      "step": 1215
    },
    {
      "entropy": 6.487839126586914,
      "epoch": 0.12107979356887653,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499994648038667,
      "loss": 6.2708,
      "mean_token_accuracy": 0.11672815084457397,
      "num_tokens": 2685073.0,
      "step": 1220
    },
    {
      "entropy": 6.450255823135376,
      "epoch": 0.12157602223104406,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999944008681496,
      "loss": 6.1999,
      "mean_token_accuracy": 0.11382194608449936,
      "num_tokens": 2695024.0,
      "step": 1225
    },
    {
      "entropy": 6.476427698135376,
      "epoch": 0.1220722508932116,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999941481182559,
      "loss": 6.2898,
      "mean_token_accuracy": 0.10765945836901665,
      "num_tokens": 2706232.0,
      "step": 1230
    },
    {
      "entropy": 6.459441423416138,
      "epoch": 0.12256847955537911,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999938897889921,
      "loss": 6.2763,
      "mean_token_accuracy": 0.11456048339605332,
      "num_tokens": 2715874.0,
      "step": 1235
    },
    {
      "entropy": 6.38568229675293,
      "epoch": 0.12306470821754664,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999936258803647,
      "loss": 6.2288,
      "mean_token_accuracy": 0.11625665873289108,
      "num_tokens": 2727409.0,
      "step": 1240
    },
    {
      "entropy": 6.370026016235352,
      "epoch": 0.12356093687971417,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999933563923802,
      "loss": 6.1212,
      "mean_token_accuracy": 0.11923045217990876,
      "num_tokens": 2737259.0,
      "step": 1245
    },
    {
      "entropy": 6.439346504211426,
      "epoch": 0.1240571655418817,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999930813250451,
      "loss": 6.2589,
      "mean_token_accuracy": 0.11177127435803413,
      "num_tokens": 2748416.0,
      "step": 1250
    },
    {
      "entropy": 6.4149510860443115,
      "epoch": 0.12455339420404922,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999928006783667,
      "loss": 6.234,
      "mean_token_accuracy": 0.11361540555953979,
      "num_tokens": 2760470.0,
      "step": 1255
    },
    {
      "entropy": 6.43356351852417,
      "epoch": 0.12504962286621676,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999925144523515,
      "loss": 6.2615,
      "mean_token_accuracy": 0.11003044992685318,
      "num_tokens": 2770222.0,
      "step": 1260
    },
    {
      "entropy": 6.444881439208984,
      "epoch": 0.12554585152838427,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999922226470066,
      "loss": 6.1815,
      "mean_token_accuracy": 0.1138396754860878,
      "num_tokens": 2780212.0,
      "step": 1265
    },
    {
      "entropy": 6.312497234344482,
      "epoch": 0.1260420801905518,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999919252623397,
      "loss": 6.1511,
      "mean_token_accuracy": 0.11787962689995765,
      "num_tokens": 2790239.0,
      "step": 1270
    },
    {
      "entropy": 6.418433570861817,
      "epoch": 0.12653830885271933,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999916222983577,
      "loss": 6.1859,
      "mean_token_accuracy": 0.11903090998530388,
      "num_tokens": 2801902.0,
      "step": 1275
    },
    {
      "entropy": 6.350197839736938,
      "epoch": 0.12703453751488686,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999913137550682,
      "loss": 6.1833,
      "mean_token_accuracy": 0.11472797095775604,
      "num_tokens": 2813268.0,
      "step": 1280
    },
    {
      "entropy": 6.413725900650024,
      "epoch": 0.1275307661770544,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999909996324791,
      "loss": 6.2532,
      "mean_token_accuracy": 0.11241401582956315,
      "num_tokens": 2824506.0,
      "step": 1285
    },
    {
      "entropy": 6.40827522277832,
      "epoch": 0.12802699483922192,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499990679930598,
      "loss": 6.1575,
      "mean_token_accuracy": 0.11616934686899186,
      "num_tokens": 2835144.0,
      "step": 1290
    },
    {
      "entropy": 6.333975553512573,
      "epoch": 0.12852322350138945,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999903546494327,
      "loss": 6.1793,
      "mean_token_accuracy": 0.11346072256565094,
      "num_tokens": 2845382.0,
      "step": 1295
    },
    {
      "entropy": 6.406157112121582,
      "epoch": 0.12901945216355698,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999900237889916,
      "loss": 6.1977,
      "mean_token_accuracy": 0.11914140731096268,
      "num_tokens": 2854697.0,
      "step": 1300
    },
    {
      "entropy": 6.399594116210937,
      "epoch": 0.12951568082572448,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999896873492827,
      "loss": 6.1988,
      "mean_token_accuracy": 0.11011647656559945,
      "num_tokens": 2866216.0,
      "step": 1305
    },
    {
      "entropy": 6.361563110351563,
      "epoch": 0.130011909487892,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999893453303143,
      "loss": 6.1669,
      "mean_token_accuracy": 0.11769109070301056,
      "num_tokens": 2877734.0,
      "step": 1310
    },
    {
      "entropy": 6.433618021011353,
      "epoch": 0.13050813815005954,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999889977320949,
      "loss": 6.2305,
      "mean_token_accuracy": 0.11750640273094178,
      "num_tokens": 2889445.0,
      "step": 1315
    },
    {
      "entropy": 6.24367208480835,
      "epoch": 0.13100436681222707,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999886445546332,
      "loss": 6.1673,
      "mean_token_accuracy": 0.1189102478325367,
      "num_tokens": 2900485.0,
      "step": 1320
    },
    {
      "entropy": 6.488619422912597,
      "epoch": 0.1315005954743946,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499988285797938,
      "loss": 6.1406,
      "mean_token_accuracy": 0.12021579965949059,
      "num_tokens": 2909795.0,
      "step": 1325
    },
    {
      "entropy": 6.218992137908936,
      "epoch": 0.13199682413656214,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499987921462018,
      "loss": 6.1739,
      "mean_token_accuracy": 0.12471199110150337,
      "num_tokens": 2920153.0,
      "step": 1330
    },
    {
      "entropy": 6.459899282455444,
      "epoch": 0.13249305279872967,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999875515468825,
      "loss": 6.162,
      "mean_token_accuracy": 0.11742881909012795,
      "num_tokens": 2930457.0,
      "step": 1335
    },
    {
      "entropy": 6.303973865509033,
      "epoch": 0.13298928146089717,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999871760525403,
      "loss": 6.2265,
      "mean_token_accuracy": 0.11418904587626458,
      "num_tokens": 2941305.0,
      "step": 1340
    },
    {
      "entropy": 6.43122968673706,
      "epoch": 0.1334855101230647,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999867949790011,
      "loss": 6.1693,
      "mean_token_accuracy": 0.11784612089395523,
      "num_tokens": 2951798.0,
      "step": 1345
    },
    {
      "entropy": 6.290275287628174,
      "epoch": 0.13398173878523223,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999864083262743,
      "loss": 6.067,
      "mean_token_accuracy": 0.12247526496648789,
      "num_tokens": 2962486.0,
      "step": 1350
    },
    {
      "entropy": 6.344256925582886,
      "epoch": 0.13447796744739976,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999860160943691,
      "loss": 6.1817,
      "mean_token_accuracy": 0.11753400415182114,
      "num_tokens": 2974317.0,
      "step": 1355
    },
    {
      "entropy": 6.320338010787964,
      "epoch": 0.1349741961095673,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999856182832958,
      "loss": 6.0902,
      "mean_token_accuracy": 0.12005901709198952,
      "num_tokens": 2985440.0,
      "step": 1360
    },
    {
      "entropy": 6.327242660522461,
      "epoch": 0.13547042477173482,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999852148930637,
      "loss": 6.1637,
      "mean_token_accuracy": 0.11592641919851303,
      "num_tokens": 2997044.0,
      "step": 1365
    },
    {
      "entropy": 6.361653184890747,
      "epoch": 0.13596665343390235,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999848059236832,
      "loss": 6.1792,
      "mean_token_accuracy": 0.11417617872357369,
      "num_tokens": 3005965.0,
      "step": 1370
    },
    {
      "entropy": 6.298781776428223,
      "epoch": 0.13646288209606988,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999843913751643,
      "loss": 6.0483,
      "mean_token_accuracy": 0.11773452535271645,
      "num_tokens": 3015031.0,
      "step": 1375
    },
    {
      "entropy": 6.287301540374756,
      "epoch": 0.13695911075823738,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999839712475174,
      "loss": 6.1151,
      "mean_token_accuracy": 0.1147402547299862,
      "num_tokens": 3026275.0,
      "step": 1380
    },
    {
      "entropy": 6.302942562103271,
      "epoch": 0.13745533942040492,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999835455407526,
      "loss": 6.1485,
      "mean_token_accuracy": 0.11648812666535377,
      "num_tokens": 3036778.0,
      "step": 1385
    },
    {
      "entropy": 6.41059980392456,
      "epoch": 0.13795156808257245,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999831142548808,
      "loss": 6.2536,
      "mean_token_accuracy": 0.11463667005300522,
      "num_tokens": 3048149.0,
      "step": 1390
    },
    {
      "entropy": 6.343495893478393,
      "epoch": 0.13844779674473998,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999826773899124,
      "loss": 6.1206,
      "mean_token_accuracy": 0.11659936383366584,
      "num_tokens": 3059149.0,
      "step": 1395
    },
    {
      "entropy": 6.2959178447723385,
      "epoch": 0.1389440254069075,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004999822349458586,
      "loss": 6.1424,
      "mean_token_accuracy": 0.12055872678756714,
      "num_tokens": 3071613.0,
      "step": 1400
    },
    {
      "entropy": 6.433216142654419,
      "epoch": 0.13944025406907504,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999817869227301,
      "loss": 6.2831,
      "mean_token_accuracy": 0.11233281642198563,
      "num_tokens": 3084742.0,
      "step": 1405
    },
    {
      "entropy": 6.355742263793945,
      "epoch": 0.13993648273124257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999813333205381,
      "loss": 6.1927,
      "mean_token_accuracy": 0.11989191100001335,
      "num_tokens": 3095356.0,
      "step": 1410
    },
    {
      "entropy": 6.345205307006836,
      "epoch": 0.14043271139341007,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999808741392937,
      "loss": 6.1828,
      "mean_token_accuracy": 0.11414567157626151,
      "num_tokens": 3107717.0,
      "step": 1415
    },
    {
      "entropy": 6.309127426147461,
      "epoch": 0.1409289400555776,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999804093790086,
      "loss": 6.1671,
      "mean_token_accuracy": 0.11339711248874665,
      "num_tokens": 3119205.0,
      "step": 1420
    },
    {
      "entropy": 6.414076328277588,
      "epoch": 0.14142516871774513,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499979939039694,
      "loss": 6.2164,
      "mean_token_accuracy": 0.11629738509655,
      "num_tokens": 3130564.0,
      "step": 1425
    },
    {
      "entropy": 6.257117033004761,
      "epoch": 0.14192139737991266,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999794631213618,
      "loss": 6.1496,
      "mean_token_accuracy": 0.11587404683232308,
      "num_tokens": 3141883.0,
      "step": 1430
    },
    {
      "entropy": 6.4088245868682865,
      "epoch": 0.1424176260420802,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999789816240236,
      "loss": 6.1524,
      "mean_token_accuracy": 0.11169286966323852,
      "num_tokens": 3152817.0,
      "step": 1435
    },
    {
      "entropy": 6.226463317871094,
      "epoch": 0.14291385470424772,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999784945476915,
      "loss": 6.0639,
      "mean_token_accuracy": 0.12211388051509857,
      "num_tokens": 3163546.0,
      "step": 1440
    },
    {
      "entropy": 6.368065643310547,
      "epoch": 0.14341008336641525,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999780018923775,
      "loss": 6.1194,
      "mean_token_accuracy": 0.11407855600118637,
      "num_tokens": 3173899.0,
      "step": 1445
    },
    {
      "entropy": 6.161925745010376,
      "epoch": 0.14390631202858278,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999775036580939,
      "loss": 6.0337,
      "mean_token_accuracy": 0.11928667798638344,
      "num_tokens": 3184433.0,
      "step": 1450
    },
    {
      "entropy": 6.412479734420776,
      "epoch": 0.1444025406907503,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999769998448529,
      "loss": 6.1516,
      "mean_token_accuracy": 0.11415581554174423,
      "num_tokens": 3196077.0,
      "step": 1455
    },
    {
      "entropy": 6.141682243347168,
      "epoch": 0.14489876935291782,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999764904526672,
      "loss": 6.0621,
      "mean_token_accuracy": 0.12485457435250283,
      "num_tokens": 3206645.0,
      "step": 1460
    },
    {
      "entropy": 6.4062056064605715,
      "epoch": 0.14539499801508535,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999759754815492,
      "loss": 6.185,
      "mean_token_accuracy": 0.12362410277128219,
      "num_tokens": 3216786.0,
      "step": 1465
    },
    {
      "entropy": 6.270987462997437,
      "epoch": 0.14589122667725288,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999754549315118,
      "loss": 6.1125,
      "mean_token_accuracy": 0.1173293948173523,
      "num_tokens": 3227384.0,
      "step": 1470
    },
    {
      "entropy": 6.215911531448365,
      "epoch": 0.1463874553394204,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499974928802568,
      "loss": 5.9691,
      "mean_token_accuracy": 0.12435413226485252,
      "num_tokens": 3238800.0,
      "step": 1475
    },
    {
      "entropy": 6.228160572052002,
      "epoch": 0.14688368400158794,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999743970947306,
      "loss": 6.1209,
      "mean_token_accuracy": 0.11474767550826073,
      "num_tokens": 3248627.0,
      "step": 1480
    },
    {
      "entropy": 6.310603427886963,
      "epoch": 0.14737991266375547,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499973859808013,
      "loss": 6.1474,
      "mean_token_accuracy": 0.11879593282938003,
      "num_tokens": 3258989.0,
      "step": 1485
    },
    {
      "entropy": 6.224552726745605,
      "epoch": 0.14787614132592297,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999733169424284,
      "loss": 6.1196,
      "mean_token_accuracy": 0.12131216377019882,
      "num_tokens": 3269474.0,
      "step": 1490
    },
    {
      "entropy": 6.298264455795288,
      "epoch": 0.1483723699880905,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999727684979904,
      "loss": 6.1048,
      "mean_token_accuracy": 0.11965034380555153,
      "num_tokens": 3280453.0,
      "step": 1495
    },
    {
      "entropy": 6.228455209732056,
      "epoch": 0.14886859865025803,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999722144747125,
      "loss": 6.0331,
      "mean_token_accuracy": 0.11981930211186409,
      "num_tokens": 3290234.0,
      "step": 1500
    },
    {
      "entropy": 6.207645463943481,
      "epoch": 0.14936482731242556,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999716548726085,
      "loss": 6.0146,
      "mean_token_accuracy": 0.12176985815167427,
      "num_tokens": 3300943.0,
      "step": 1505
    },
    {
      "entropy": 6.184590911865234,
      "epoch": 0.1498610559745931,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999710896916921,
      "loss": 5.9902,
      "mean_token_accuracy": 0.12336606532335281,
      "num_tokens": 3311296.0,
      "step": 1510
    },
    {
      "entropy": 6.256126356124878,
      "epoch": 0.15035728463676062,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999705189319774,
      "loss": 6.1279,
      "mean_token_accuracy": 0.11304511949419975,
      "num_tokens": 3322310.0,
      "step": 1515
    },
    {
      "entropy": 6.240859413146973,
      "epoch": 0.15085351329892815,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999699425934788,
      "loss": 6.1363,
      "mean_token_accuracy": 0.11859385669231415,
      "num_tokens": 3334176.0,
      "step": 1520
    },
    {
      "entropy": 6.342157602310181,
      "epoch": 0.15134974196109569,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999693606762103,
      "loss": 6.1278,
      "mean_token_accuracy": 0.11730345636606217,
      "num_tokens": 3344839.0,
      "step": 1525
    },
    {
      "entropy": 6.24994010925293,
      "epoch": 0.1518459706232632,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999687731801865,
      "loss": 6.1041,
      "mean_token_accuracy": 0.11314293071627617,
      "num_tokens": 3357046.0,
      "step": 1530
    },
    {
      "entropy": 6.32585072517395,
      "epoch": 0.15234219928543072,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999681801054217,
      "loss": 6.1754,
      "mean_token_accuracy": 0.11708802953362465,
      "num_tokens": 3367783.0,
      "step": 1535
    },
    {
      "entropy": 6.213843870162964,
      "epoch": 0.15283842794759825,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499967581451931,
      "loss": 6.0432,
      "mean_token_accuracy": 0.11833713948726654,
      "num_tokens": 3378529.0,
      "step": 1540
    },
    {
      "entropy": 6.292020654678344,
      "epoch": 0.15333465660976578,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499966977219729,
      "loss": 6.2356,
      "mean_token_accuracy": 0.11526345536112785,
      "num_tokens": 3389318.0,
      "step": 1545
    },
    {
      "entropy": 6.392855215072632,
      "epoch": 0.1538308852719333,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999663674088306,
      "loss": 6.2284,
      "mean_token_accuracy": 0.11469097509980201,
      "num_tokens": 3400603.0,
      "step": 1550
    },
    {
      "entropy": 6.286278486251831,
      "epoch": 0.15432711393410084,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999657520192511,
      "loss": 6.1222,
      "mean_token_accuracy": 0.11655379831790924,
      "num_tokens": 3410316.0,
      "step": 1555
    },
    {
      "entropy": 6.2597517490386965,
      "epoch": 0.15482334259626837,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999651310510058,
      "loss": 6.0634,
      "mean_token_accuracy": 0.12244669795036316,
      "num_tokens": 3422155.0,
      "step": 1560
    },
    {
      "entropy": 6.282853603363037,
      "epoch": 0.15531957125843587,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999645045041099,
      "loss": 6.0028,
      "mean_token_accuracy": 0.1191568560898304,
      "num_tokens": 3432892.0,
      "step": 1565
    },
    {
      "entropy": 6.221517276763916,
      "epoch": 0.1558157999206034,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999638723785791,
      "loss": 6.1304,
      "mean_token_accuracy": 0.12089115679264069,
      "num_tokens": 3443967.0,
      "step": 1570
    },
    {
      "entropy": 6.202426433563232,
      "epoch": 0.15631202858277093,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499963234674429,
      "loss": 6.0213,
      "mean_token_accuracy": 0.12375078648328781,
      "num_tokens": 3454354.0,
      "step": 1575
    },
    {
      "entropy": 6.131307458877563,
      "epoch": 0.15680825724493846,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999625913916754,
      "loss": 6.0405,
      "mean_token_accuracy": 0.11991182640194893,
      "num_tokens": 3465232.0,
      "step": 1580
    },
    {
      "entropy": 6.1873551368713375,
      "epoch": 0.157304485907106,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999619425303343,
      "loss": 6.0184,
      "mean_token_accuracy": 0.12445951774716377,
      "num_tokens": 3475443.0,
      "step": 1585
    },
    {
      "entropy": 6.2718860626220705,
      "epoch": 0.15780071456927353,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999612880904219,
      "loss": 6.0401,
      "mean_token_accuracy": 0.12001754567027093,
      "num_tokens": 3486643.0,
      "step": 1590
    },
    {
      "entropy": 6.218065166473389,
      "epoch": 0.15829694323144106,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999606280719542,
      "loss": 6.1047,
      "mean_token_accuracy": 0.1197525791823864,
      "num_tokens": 3497557.0,
      "step": 1595
    },
    {
      "entropy": 6.363813638687134,
      "epoch": 0.1587931718936086,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999599624749475,
      "loss": 6.1079,
      "mean_token_accuracy": 0.1160752110183239,
      "num_tokens": 3509790.0,
      "step": 1600
    },
    {
      "entropy": 6.141760301589966,
      "epoch": 0.1592894005557761,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999592912994187,
      "loss": 6.0233,
      "mean_token_accuracy": 0.11927468180656434,
      "num_tokens": 3521577.0,
      "step": 1605
    },
    {
      "entropy": 6.263496160507202,
      "epoch": 0.15978562921794362,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999586145453842,
      "loss": 6.0717,
      "mean_token_accuracy": 0.11626604050397873,
      "num_tokens": 3532126.0,
      "step": 1610
    },
    {
      "entropy": 6.084614086151123,
      "epoch": 0.16028185788011115,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999579322128608,
      "loss": 5.9055,
      "mean_token_accuracy": 0.12044439241290092,
      "num_tokens": 3542560.0,
      "step": 1615
    },
    {
      "entropy": 6.201308965682983,
      "epoch": 0.16077808654227868,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999572443018654,
      "loss": 6.065,
      "mean_token_accuracy": 0.11997904106974602,
      "num_tokens": 3553863.0,
      "step": 1620
    },
    {
      "entropy": 6.1602541446685795,
      "epoch": 0.1612743152044462,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999565508124151,
      "loss": 6.0874,
      "mean_token_accuracy": 0.12250801846385002,
      "num_tokens": 3564953.0,
      "step": 1625
    },
    {
      "entropy": 6.266576147079467,
      "epoch": 0.16177054386661374,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499955851744527,
      "loss": 6.0789,
      "mean_token_accuracy": 0.12218151539564133,
      "num_tokens": 3576247.0,
      "step": 1630
    },
    {
      "entropy": 6.306499862670899,
      "epoch": 0.16226677252878127,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999551470982185,
      "loss": 6.1101,
      "mean_token_accuracy": 0.11723152995109558,
      "num_tokens": 3587600.0,
      "step": 1635
    },
    {
      "entropy": 6.263530349731445,
      "epoch": 0.1627630011909488,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999544368735073,
      "loss": 6.0821,
      "mean_token_accuracy": 0.12516205981373787,
      "num_tokens": 3597727.0,
      "step": 1640
    },
    {
      "entropy": 6.230346632003784,
      "epoch": 0.1632592298531163,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999537210704106,
      "loss": 6.0679,
      "mean_token_accuracy": 0.12371558174490929,
      "num_tokens": 3608066.0,
      "step": 1645
    },
    {
      "entropy": 6.272209596633911,
      "epoch": 0.16375545851528384,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999529996889463,
      "loss": 6.1215,
      "mean_token_accuracy": 0.11630801856517792,
      "num_tokens": 3618263.0,
      "step": 1650
    },
    {
      "entropy": 6.252897644042969,
      "epoch": 0.16425168717745137,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999522727291325,
      "loss": 6.07,
      "mean_token_accuracy": 0.12269503250718117,
      "num_tokens": 3629465.0,
      "step": 1655
    },
    {
      "entropy": 6.205037641525268,
      "epoch": 0.1647479158396189,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499951540190987,
      "loss": 6.0946,
      "mean_token_accuracy": 0.11775854378938674,
      "num_tokens": 3639657.0,
      "step": 1660
    },
    {
      "entropy": 6.27462944984436,
      "epoch": 0.16524414450178643,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999508020745281,
      "loss": 6.0516,
      "mean_token_accuracy": 0.11680529490113259,
      "num_tokens": 3652374.0,
      "step": 1665
    },
    {
      "entropy": 6.219149589538574,
      "epoch": 0.16574037316395396,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999500583797739,
      "loss": 6.0272,
      "mean_token_accuracy": 0.12123104184865952,
      "num_tokens": 3664083.0,
      "step": 1670
    },
    {
      "entropy": 6.1698380470275875,
      "epoch": 0.1662366018261215,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999493091067431,
      "loss": 6.0679,
      "mean_token_accuracy": 0.12645655497908592,
      "num_tokens": 3675314.0,
      "step": 1675
    },
    {
      "entropy": 6.184545946121216,
      "epoch": 0.166732830488289,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999485542554541,
      "loss": 6.0337,
      "mean_token_accuracy": 0.11284921467304229,
      "num_tokens": 3686019.0,
      "step": 1680
    },
    {
      "entropy": 6.288911151885986,
      "epoch": 0.16722905915045652,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999477938259256,
      "loss": 6.1241,
      "mean_token_accuracy": 0.1167833961546421,
      "num_tokens": 3698170.0,
      "step": 1685
    },
    {
      "entropy": 6.210192346572876,
      "epoch": 0.16772528781262405,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999470278181766,
      "loss": 6.1146,
      "mean_token_accuracy": 0.11569419503211975,
      "num_tokens": 3709243.0,
      "step": 1690
    },
    {
      "entropy": 6.2869891166687015,
      "epoch": 0.16822151647479158,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999462562322261,
      "loss": 6.0797,
      "mean_token_accuracy": 0.12025846838951111,
      "num_tokens": 3720196.0,
      "step": 1695
    },
    {
      "entropy": 6.290680551528931,
      "epoch": 0.1687177451369591,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499945479068093,
      "loss": 6.1355,
      "mean_token_accuracy": 0.12103915438055993,
      "num_tokens": 3732714.0,
      "step": 1700
    },
    {
      "entropy": 6.222851324081421,
      "epoch": 0.16921397379912664,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004999446963257969,
      "loss": 6.0964,
      "mean_token_accuracy": 0.11665565371513367,
      "num_tokens": 3746474.0,
      "step": 1705
    },
    {
      "entropy": 6.142470836639404,
      "epoch": 0.16971020246129417,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999439080053571,
      "loss": 5.9737,
      "mean_token_accuracy": 0.12647735849022865,
      "num_tokens": 3757528.0,
      "step": 1710
    },
    {
      "entropy": 6.312270736694336,
      "epoch": 0.1702064311234617,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999431141067928,
      "loss": 6.1028,
      "mean_token_accuracy": 0.1200647495687008,
      "num_tokens": 3768360.0,
      "step": 1715
    },
    {
      "entropy": 6.174591398239135,
      "epoch": 0.1707026597856292,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999423146301242,
      "loss": 5.9725,
      "mean_token_accuracy": 0.1192103810608387,
      "num_tokens": 3779809.0,
      "step": 1720
    },
    {
      "entropy": 6.11200361251831,
      "epoch": 0.17119888844779674,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999415095753709,
      "loss": 5.9783,
      "mean_token_accuracy": 0.12067282646894455,
      "num_tokens": 3791455.0,
      "step": 1725
    },
    {
      "entropy": 6.143737411499023,
      "epoch": 0.17169511710996427,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999406989425527,
      "loss": 5.9996,
      "mean_token_accuracy": 0.12016435712575912,
      "num_tokens": 3803209.0,
      "step": 1730
    },
    {
      "entropy": 6.139561700820923,
      "epoch": 0.1721913457721318,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999398827316901,
      "loss": 6.0474,
      "mean_token_accuracy": 0.12039401978254319,
      "num_tokens": 3813553.0,
      "step": 1735
    },
    {
      "entropy": 6.202684926986694,
      "epoch": 0.17268757443429933,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499939060942803,
      "loss": 6.0878,
      "mean_token_accuracy": 0.11598005071282387,
      "num_tokens": 3824856.0,
      "step": 1740
    },
    {
      "entropy": 6.242949295043945,
      "epoch": 0.17318380309646686,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999382335759119,
      "loss": 6.1542,
      "mean_token_accuracy": 0.11370768472552299,
      "num_tokens": 3835489.0,
      "step": 1745
    },
    {
      "entropy": 6.265321969985962,
      "epoch": 0.1736800317586344,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999374006310372,
      "loss": 6.0015,
      "mean_token_accuracy": 0.12164871916174888,
      "num_tokens": 3844838.0,
      "step": 1750
    },
    {
      "entropy": 6.10846905708313,
      "epoch": 0.1741762604208019,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999365621081999,
      "loss": 5.9625,
      "mean_token_accuracy": 0.11757483780384063,
      "num_tokens": 3855288.0,
      "step": 1755
    },
    {
      "entropy": 6.1842241287231445,
      "epoch": 0.17467248908296942,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999357180074204,
      "loss": 6.0165,
      "mean_token_accuracy": 0.12564777955412865,
      "num_tokens": 3865585.0,
      "step": 1760
    },
    {
      "entropy": 6.159081268310547,
      "epoch": 0.17516871774513695,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999348683287199,
      "loss": 5.9979,
      "mean_token_accuracy": 0.11737576574087143,
      "num_tokens": 3877694.0,
      "step": 1765
    },
    {
      "entropy": 6.146845006942749,
      "epoch": 0.17566494640730448,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999340130721192,
      "loss": 5.9587,
      "mean_token_accuracy": 0.13080541640520096,
      "num_tokens": 3888802.0,
      "step": 1770
    },
    {
      "entropy": 6.100156831741333,
      "epoch": 0.17616117506947201,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999331522376396,
      "loss": 5.9332,
      "mean_token_accuracy": 0.12021833211183548,
      "num_tokens": 3900294.0,
      "step": 1775
    },
    {
      "entropy": 6.149476766586304,
      "epoch": 0.17665740373163955,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999322858253027,
      "loss": 5.9754,
      "mean_token_accuracy": 0.12123704552650452,
      "num_tokens": 3910865.0,
      "step": 1780
    },
    {
      "entropy": 6.165459203720093,
      "epoch": 0.17715363239380708,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999314138351297,
      "loss": 6.0263,
      "mean_token_accuracy": 0.12229800149798394,
      "num_tokens": 3921960.0,
      "step": 1785
    },
    {
      "entropy": 6.191252756118774,
      "epoch": 0.1776498610559746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999305362671424,
      "loss": 6.0387,
      "mean_token_accuracy": 0.11913010105490685,
      "num_tokens": 3934085.0,
      "step": 1790
    },
    {
      "entropy": 6.127382612228393,
      "epoch": 0.1781460897181421,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999296531213624,
      "loss": 5.8928,
      "mean_token_accuracy": 0.13389062657952308,
      "num_tokens": 3944532.0,
      "step": 1795
    },
    {
      "entropy": 6.108739185333252,
      "epoch": 0.17864231838030964,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999287643978117,
      "loss": 5.9284,
      "mean_token_accuracy": 0.1266961745917797,
      "num_tokens": 3954995.0,
      "step": 1800
    },
    {
      "entropy": 6.063549137115478,
      "epoch": 0.17913854704247717,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999278700965124,
      "loss": 5.99,
      "mean_token_accuracy": 0.1248068444430828,
      "num_tokens": 3966621.0,
      "step": 1805
    },
    {
      "entropy": 6.155789041519165,
      "epoch": 0.1796347757046447,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999269702174866,
      "loss": 6.0518,
      "mean_token_accuracy": 0.12221262827515603,
      "num_tokens": 3977568.0,
      "step": 1810
    },
    {
      "entropy": 6.161475467681885,
      "epoch": 0.18013100436681223,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999260647607565,
      "loss": 6.0683,
      "mean_token_accuracy": 0.11906974241137505,
      "num_tokens": 3988571.0,
      "step": 1815
    },
    {
      "entropy": 6.180405378341675,
      "epoch": 0.18062723302897976,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999251537263448,
      "loss": 5.8891,
      "mean_token_accuracy": 0.12807116732001306,
      "num_tokens": 3998380.0,
      "step": 1820
    },
    {
      "entropy": 6.073153924942017,
      "epoch": 0.1811234616911473,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999242371142739,
      "loss": 6.0713,
      "mean_token_accuracy": 0.12228276878595352,
      "num_tokens": 4010017.0,
      "step": 1825
    },
    {
      "entropy": 6.301033782958984,
      "epoch": 0.1816196903533148,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999233149245665,
      "loss": 6.0383,
      "mean_token_accuracy": 0.12104798778891564,
      "num_tokens": 4020922.0,
      "step": 1830
    },
    {
      "entropy": 6.202951669692993,
      "epoch": 0.18211591901548232,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999223871572456,
      "loss": 6.0035,
      "mean_token_accuracy": 0.12094608992338181,
      "num_tokens": 4030927.0,
      "step": 1835
    },
    {
      "entropy": 6.1489091396331785,
      "epoch": 0.18261214767764986,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999214538123342,
      "loss": 6.0733,
      "mean_token_accuracy": 0.11611957773566246,
      "num_tokens": 4041375.0,
      "step": 1840
    },
    {
      "entropy": 6.178863096237182,
      "epoch": 0.18310837633981739,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999205148898554,
      "loss": 6.0746,
      "mean_token_accuracy": 0.11968084275722504,
      "num_tokens": 4053213.0,
      "step": 1845
    },
    {
      "entropy": 6.172112369537354,
      "epoch": 0.18360460500198492,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999195703898325,
      "loss": 5.9206,
      "mean_token_accuracy": 0.12607544884085656,
      "num_tokens": 4064536.0,
      "step": 1850
    },
    {
      "entropy": 6.156340265274048,
      "epoch": 0.18410083366415245,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999186203122889,
      "loss": 5.9853,
      "mean_token_accuracy": 0.12510076612234117,
      "num_tokens": 4075282.0,
      "step": 1855
    },
    {
      "entropy": 6.041736364364624,
      "epoch": 0.18459706232631998,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999176646572482,
      "loss": 5.9996,
      "mean_token_accuracy": 0.12671509012579918,
      "num_tokens": 4085501.0,
      "step": 1860
    },
    {
      "entropy": 6.163187837600708,
      "epoch": 0.1850932909884875,
      "grad_norm": 1.0,
      "learning_rate": 0.000499916703424734,
      "loss": 5.9386,
      "mean_token_accuracy": 0.11762271374464035,
      "num_tokens": 4095697.0,
      "step": 1865
    },
    {
      "entropy": 6.143773317337036,
      "epoch": 0.185589519650655,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999157366147702,
      "loss": 6.0013,
      "mean_token_accuracy": 0.12127534747123718,
      "num_tokens": 4106459.0,
      "step": 1870
    },
    {
      "entropy": 6.127069616317749,
      "epoch": 0.18608574831282254,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999147642273809,
      "loss": 5.9328,
      "mean_token_accuracy": 0.12429732158780098,
      "num_tokens": 4118320.0,
      "step": 1875
    },
    {
      "entropy": 6.075209379196167,
      "epoch": 0.18658197697499007,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049991378626259,
      "loss": 5.9491,
      "mean_token_accuracy": 0.13264545872807504,
      "num_tokens": 4131249.0,
      "step": 1880
    },
    {
      "entropy": 6.168709850311279,
      "epoch": 0.1870782056371576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999128027204219,
      "loss": 6.0205,
      "mean_token_accuracy": 0.1249746061861515,
      "num_tokens": 4141480.0,
      "step": 1885
    },
    {
      "entropy": 6.181358861923218,
      "epoch": 0.18757443429932513,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499911813600901,
      "loss": 6.0712,
      "mean_token_accuracy": 0.12171259969472885,
      "num_tokens": 4152344.0,
      "step": 1890
    },
    {
      "entropy": 6.18400239944458,
      "epoch": 0.18807066296149266,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999108189040516,
      "loss": 6.0532,
      "mean_token_accuracy": 0.1188795156776905,
      "num_tokens": 4164772.0,
      "step": 1895
    },
    {
      "entropy": 6.120943212509156,
      "epoch": 0.1885668916236602,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999098186298986,
      "loss": 5.9959,
      "mean_token_accuracy": 0.11978129372000694,
      "num_tokens": 4175511.0,
      "step": 1900
    },
    {
      "entropy": 6.083007574081421,
      "epoch": 0.1890631202858277,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999088127784667,
      "loss": 5.9597,
      "mean_token_accuracy": 0.1269135594367981,
      "num_tokens": 4187445.0,
      "step": 1905
    },
    {
      "entropy": 6.157655334472656,
      "epoch": 0.18955934894799523,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499907801349781,
      "loss": 5.9288,
      "mean_token_accuracy": 0.1249204121530056,
      "num_tokens": 4198124.0,
      "step": 1910
    },
    {
      "entropy": 6.117712736129761,
      "epoch": 0.19005557761016276,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999067843438664,
      "loss": 5.9642,
      "mean_token_accuracy": 0.12320106327533722,
      "num_tokens": 4208954.0,
      "step": 1915
    },
    {
      "entropy": 6.1655356884002686,
      "epoch": 0.1905518062723303,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999057617607482,
      "loss": 6.036,
      "mean_token_accuracy": 0.12222218587994575,
      "num_tokens": 4218896.0,
      "step": 1920
    },
    {
      "entropy": 6.073115491867066,
      "epoch": 0.19104803493449782,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999047336004517,
      "loss": 5.9545,
      "mean_token_accuracy": 0.11575927585363388,
      "num_tokens": 4228851.0,
      "step": 1925
    },
    {
      "entropy": 6.062970113754273,
      "epoch": 0.19154426359666535,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999036998630024,
      "loss": 5.9034,
      "mean_token_accuracy": 0.12176035344600677,
      "num_tokens": 4240377.0,
      "step": 1930
    },
    {
      "entropy": 6.1012085437774655,
      "epoch": 0.19204049225883288,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499902660548426,
      "loss": 5.9005,
      "mean_token_accuracy": 0.12233036682009697,
      "num_tokens": 4251341.0,
      "step": 1935
    },
    {
      "entropy": 6.09600863456726,
      "epoch": 0.1925367209210004,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999016156567482,
      "loss": 5.9999,
      "mean_token_accuracy": 0.12348600029945374,
      "num_tokens": 4261206.0,
      "step": 1940
    },
    {
      "entropy": 6.171257972717285,
      "epoch": 0.1930329495831679,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499900565187995,
      "loss": 6.0095,
      "mean_token_accuracy": 0.11739006116986275,
      "num_tokens": 4272662.0,
      "step": 1945
    },
    {
      "entropy": 6.155332422256469,
      "epoch": 0.19352917824533544,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998995091421924,
      "loss": 5.9894,
      "mean_token_accuracy": 0.12250523716211319,
      "num_tokens": 4283302.0,
      "step": 1950
    },
    {
      "entropy": 6.102492046356201,
      "epoch": 0.19402540690750297,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998984475193665,
      "loss": 5.9331,
      "mean_token_accuracy": 0.1246795579791069,
      "num_tokens": 4296087.0,
      "step": 1955
    },
    {
      "entropy": 6.071946620941162,
      "epoch": 0.1945216355696705,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998973803195439,
      "loss": 5.9506,
      "mean_token_accuracy": 0.12343922182917595,
      "num_tokens": 4305918.0,
      "step": 1960
    },
    {
      "entropy": 6.098349571228027,
      "epoch": 0.19501786423183803,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998963075427508,
      "loss": 6.0056,
      "mean_token_accuracy": 0.12570720240473748,
      "num_tokens": 4316507.0,
      "step": 1965
    },
    {
      "entropy": 6.172418403625488,
      "epoch": 0.19551409289400556,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004998952291890138,
      "loss": 5.9377,
      "mean_token_accuracy": 0.12697042599320413,
      "num_tokens": 4329091.0,
      "step": 1970
    },
    {
      "entropy": 6.116201543807984,
      "epoch": 0.1960103215561731,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998941452583599,
      "loss": 6.0005,
      "mean_token_accuracy": 0.12526401430368422,
      "num_tokens": 4339266.0,
      "step": 1975
    },
    {
      "entropy": 6.060456848144531,
      "epoch": 0.1965065502183406,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998930557508156,
      "loss": 5.9315,
      "mean_token_accuracy": 0.12996648699045182,
      "num_tokens": 4349300.0,
      "step": 1980
    },
    {
      "entropy": 6.031670236587525,
      "epoch": 0.19700277888050813,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998919606664082,
      "loss": 5.8523,
      "mean_token_accuracy": 0.1329902485013008,
      "num_tokens": 4360314.0,
      "step": 1985
    },
    {
      "entropy": 6.06371750831604,
      "epoch": 0.19749900754267566,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998908600051648,
      "loss": 5.9022,
      "mean_token_accuracy": 0.12996097579598426,
      "num_tokens": 4370077.0,
      "step": 1990
    },
    {
      "entropy": 6.041788816452026,
      "epoch": 0.1979952362048432,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998897537671126,
      "loss": 5.9163,
      "mean_token_accuracy": 0.12319361120462417,
      "num_tokens": 4381000.0,
      "step": 1995
    },
    {
      "entropy": 6.162951469421387,
      "epoch": 0.19849146486701072,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998886419522792,
      "loss": 6.1233,
      "mean_token_accuracy": 0.11464894115924835,
      "num_tokens": 4393485.0,
      "step": 2000
    },
    {
      "entropy": 6.207508325576782,
      "epoch": 0.19898769352917825,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998875245606919,
      "loss": 5.9466,
      "mean_token_accuracy": 0.1263269901275635,
      "num_tokens": 4404354.0,
      "step": 2005
    },
    {
      "entropy": 6.115656423568725,
      "epoch": 0.19948392219134578,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998864015923788,
      "loss": 6.0848,
      "mean_token_accuracy": 0.12513597384095193,
      "num_tokens": 4414818.0,
      "step": 2010
    },
    {
      "entropy": 6.1054645538330075,
      "epoch": 0.1999801508535133,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998852730473674,
      "loss": 5.9732,
      "mean_token_accuracy": 0.1186623029410839,
      "num_tokens": 4426271.0,
      "step": 2015
    },
    {
      "entropy": 6.113159704208374,
      "epoch": 0.2004763795156808,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998841389256859,
      "loss": 6.0201,
      "mean_token_accuracy": 0.12535962387919425,
      "num_tokens": 4436981.0,
      "step": 2020
    },
    {
      "entropy": 6.088190221786499,
      "epoch": 0.20097260817784834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998829992273622,
      "loss": 5.8457,
      "mean_token_accuracy": 0.12926107719540597,
      "num_tokens": 4446494.0,
      "step": 2025
    },
    {
      "entropy": 6.0542316913604735,
      "epoch": 0.20146883684001587,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004998818539524248,
      "loss": 5.941,
      "mean_token_accuracy": 0.12792145758867263,
      "num_tokens": 4458535.0,
      "step": 2030
    },
    {
      "entropy": 6.076647806167602,
      "epoch": 0.2019650655021834,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499880703100902,
      "loss": 5.9109,
      "mean_token_accuracy": 0.12890672460198402,
      "num_tokens": 4469025.0,
      "step": 2035
    },
    {
      "entropy": 6.119675874710083,
      "epoch": 0.20246129416435094,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998795466728223,
      "loss": 5.9311,
      "mean_token_accuracy": 0.12614311054348945,
      "num_tokens": 4478431.0,
      "step": 2040
    },
    {
      "entropy": 6.100471878051758,
      "epoch": 0.20295752282651847,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998783846682144,
      "loss": 6.0565,
      "mean_token_accuracy": 0.12169205471873283,
      "num_tokens": 4490215.0,
      "step": 2045
    },
    {
      "entropy": 6.08721661567688,
      "epoch": 0.203453751488686,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004998772170871072,
      "loss": 5.9323,
      "mean_token_accuracy": 0.12300658226013184,
      "num_tokens": 4502895.0,
      "step": 2050
    },
    {
      "entropy": 6.049704933166504,
      "epoch": 0.20394998015085353,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998760439295295,
      "loss": 5.9527,
      "mean_token_accuracy": 0.12245631441473961,
      "num_tokens": 4514468.0,
      "step": 2055
    },
    {
      "entropy": 6.120728397369385,
      "epoch": 0.20444620881302103,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998748651955105,
      "loss": 5.9697,
      "mean_token_accuracy": 0.12270477414131165,
      "num_tokens": 4525787.0,
      "step": 2060
    },
    {
      "entropy": 6.1617701053619385,
      "epoch": 0.20494243747518856,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004998736808850795,
      "loss": 6.0537,
      "mean_token_accuracy": 0.1189173199236393,
      "num_tokens": 4537536.0,
      "step": 2065
    },
    {
      "entropy": 6.137720203399658,
      "epoch": 0.2054386661373561,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998724909982656,
      "loss": 5.9957,
      "mean_token_accuracy": 0.1230063110589981,
      "num_tokens": 4549619.0,
      "step": 2070
    },
    {
      "entropy": 6.054072999954224,
      "epoch": 0.20593489479952362,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998712955350986,
      "loss": 5.9127,
      "mean_token_accuracy": 0.12620335072278976,
      "num_tokens": 4560248.0,
      "step": 2075
    },
    {
      "entropy": 6.133942365646362,
      "epoch": 0.20643112346169115,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998700944956081,
      "loss": 6.0258,
      "mean_token_accuracy": 0.12564925700426102,
      "num_tokens": 4571888.0,
      "step": 2080
    },
    {
      "entropy": 6.092451715469361,
      "epoch": 0.20692735212385868,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998688878798237,
      "loss": 5.8719,
      "mean_token_accuracy": 0.13250108286738396,
      "num_tokens": 4583234.0,
      "step": 2085
    },
    {
      "entropy": 6.129697036743164,
      "epoch": 0.2074235807860262,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004998676756877753,
      "loss": 6.0849,
      "mean_token_accuracy": 0.1196140132844448,
      "num_tokens": 4595252.0,
      "step": 2090
    },
    {
      "entropy": 6.055380916595459,
      "epoch": 0.20791980944819372,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998664579194934,
      "loss": 5.8636,
      "mean_token_accuracy": 0.12891709432005882,
      "num_tokens": 4607091.0,
      "step": 2095
    },
    {
      "entropy": 6.060617876052857,
      "epoch": 0.20841603811036125,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004998652345750077,
      "loss": 5.8251,
      "mean_token_accuracy": 0.13188225999474526,
      "num_tokens": 4618003.0,
      "step": 2100
    },
    {
      "entropy": 6.004556941986084,
      "epoch": 0.20891226677252878,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998640056543487,
      "loss": 5.8382,
      "mean_token_accuracy": 0.12877558246254922,
      "num_tokens": 4628429.0,
      "step": 2105
    },
    {
      "entropy": 6.042101526260376,
      "epoch": 0.2094084954346963,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998627711575469,
      "loss": 5.9003,
      "mean_token_accuracy": 0.1301546387374401,
      "num_tokens": 4639414.0,
      "step": 2110
    },
    {
      "entropy": 6.125416374206543,
      "epoch": 0.20990472409686384,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000499861531084633,
      "loss": 5.9868,
      "mean_token_accuracy": 0.12510028034448623,
      "num_tokens": 4652594.0,
      "step": 2115
    },
    {
      "entropy": 5.983136796951294,
      "epoch": 0.21040095275903137,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998602854356376,
      "loss": 5.8353,
      "mean_token_accuracy": 0.13099646791815758,
      "num_tokens": 4662981.0,
      "step": 2120
    },
    {
      "entropy": 6.04475212097168,
      "epoch": 0.2108971814211989,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998590342105917,
      "loss": 5.929,
      "mean_token_accuracy": 0.12667660117149354,
      "num_tokens": 4672469.0,
      "step": 2125
    },
    {
      "entropy": 6.043806028366089,
      "epoch": 0.21139341008336643,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998577774095262,
      "loss": 5.9052,
      "mean_token_accuracy": 0.12564729601144792,
      "num_tokens": 4683640.0,
      "step": 2130
    },
    {
      "entropy": 6.0359203815460205,
      "epoch": 0.21188963874553393,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998565150324724,
      "loss": 5.8664,
      "mean_token_accuracy": 0.12986178547143937,
      "num_tokens": 4694032.0,
      "step": 2135
    },
    {
      "entropy": 5.933635091781616,
      "epoch": 0.21238586740770146,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998552470794616,
      "loss": 5.7421,
      "mean_token_accuracy": 0.13141876757144927,
      "num_tokens": 4704253.0,
      "step": 2140
    },
    {
      "entropy": 6.008824586868286,
      "epoch": 0.212882096069869,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998539735505251,
      "loss": 5.8993,
      "mean_token_accuracy": 0.1256441943347454,
      "num_tokens": 4715327.0,
      "step": 2145
    },
    {
      "entropy": 6.166449022293091,
      "epoch": 0.21337832473203652,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998526944456946,
      "loss": 5.9113,
      "mean_token_accuracy": 0.12448364496231079,
      "num_tokens": 4726450.0,
      "step": 2150
    },
    {
      "entropy": 6.0194117546081545,
      "epoch": 0.21387455339420405,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998514097650018,
      "loss": 5.9441,
      "mean_token_accuracy": 0.12404742538928985,
      "num_tokens": 4736630.0,
      "step": 2155
    },
    {
      "entropy": 6.021891641616821,
      "epoch": 0.21437078205637158,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998501195084786,
      "loss": 5.8261,
      "mean_token_accuracy": 0.13041977211833,
      "num_tokens": 4746592.0,
      "step": 2160
    },
    {
      "entropy": 6.075963544845581,
      "epoch": 0.21486701071853911,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998488236761569,
      "loss": 5.9943,
      "mean_token_accuracy": 0.12953520566225052,
      "num_tokens": 4757824.0,
      "step": 2165
    },
    {
      "entropy": 6.078512859344483,
      "epoch": 0.21536323938070662,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998475222680688,
      "loss": 5.9418,
      "mean_token_accuracy": 0.12912883386015891,
      "num_tokens": 4768453.0,
      "step": 2170
    },
    {
      "entropy": 5.984212827682495,
      "epoch": 0.21585946804287415,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998462152842468,
      "loss": 5.9138,
      "mean_token_accuracy": 0.1260770782828331,
      "num_tokens": 4779664.0,
      "step": 2175
    },
    {
      "entropy": 6.1202507495880125,
      "epoch": 0.21635569670504168,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499844902724723,
      "loss": 5.9297,
      "mean_token_accuracy": 0.1291131466627121,
      "num_tokens": 4791012.0,
      "step": 2180
    },
    {
      "entropy": 6.0246453285217285,
      "epoch": 0.2168519253672092,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998435845895302,
      "loss": 5.804,
      "mean_token_accuracy": 0.1243359960615635,
      "num_tokens": 4800578.0,
      "step": 2185
    },
    {
      "entropy": 5.987508344650268,
      "epoch": 0.21734815402937674,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499842260878701,
      "loss": 5.8239,
      "mean_token_accuracy": 0.12554812505841256,
      "num_tokens": 4810940.0,
      "step": 2190
    },
    {
      "entropy": 6.046344041824341,
      "epoch": 0.21784438269154427,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004998409315922683,
      "loss": 5.9594,
      "mean_token_accuracy": 0.12463009878993034,
      "num_tokens": 4822696.0,
      "step": 2195
    },
    {
      "entropy": 6.06123194694519,
      "epoch": 0.2183406113537118,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998395967302649,
      "loss": 5.8424,
      "mean_token_accuracy": 0.1316361606121063,
      "num_tokens": 4833669.0,
      "step": 2200
    },
    {
      "entropy": 6.1397562503814695,
      "epoch": 0.21883684001587933,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499838256292724,
      "loss": 6.0156,
      "mean_token_accuracy": 0.12383595556020736,
      "num_tokens": 4844021.0,
      "step": 2205
    },
    {
      "entropy": 5.95759687423706,
      "epoch": 0.21933306867804683,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998369102796787,
      "loss": 5.8932,
      "mean_token_accuracy": 0.12637994140386583,
      "num_tokens": 4855336.0,
      "step": 2210
    },
    {
      "entropy": 6.048965978622436,
      "epoch": 0.21982929734021436,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004998355586911627,
      "loss": 5.9059,
      "mean_token_accuracy": 0.12790409997105598,
      "num_tokens": 4867713.0,
      "step": 2215
    },
    {
      "entropy": 5.9969117641448975,
      "epoch": 0.2203255260023819,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004998342015272093,
      "loss": 5.8694,
      "mean_token_accuracy": 0.12453557699918746,
      "num_tokens": 4879134.0,
      "step": 2220
    },
    {
      "entropy": 6.086967134475708,
      "epoch": 0.22082175466454942,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004998328387878521,
      "loss": 6.0274,
      "mean_token_accuracy": 0.12505436688661575,
      "num_tokens": 4891755.0,
      "step": 2225
    },
    {
      "entropy": 6.152056741714477,
      "epoch": 0.22131798332671695,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998314704731249,
      "loss": 6.0155,
      "mean_token_accuracy": 0.12360429838299751,
      "num_tokens": 4903264.0,
      "step": 2230
    },
    {
      "entropy": 6.021547603607178,
      "epoch": 0.22181421198888449,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998300965830618,
      "loss": 5.8482,
      "mean_token_accuracy": 0.1265365608036518,
      "num_tokens": 4912533.0,
      "step": 2235
    },
    {
      "entropy": 5.9325371265411375,
      "epoch": 0.22231044065105202,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004998287171176968,
      "loss": 5.8439,
      "mean_token_accuracy": 0.13453373461961746,
      "num_tokens": 4923288.0,
      "step": 2240
    },
    {
      "entropy": 6.013333702087403,
      "epoch": 0.22280666931321952,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998273320770641,
      "loss": 5.861,
      "mean_token_accuracy": 0.1291363902390003,
      "num_tokens": 4934759.0,
      "step": 2245
    },
    {
      "entropy": 6.0565742492675785,
      "epoch": 0.22330289797538705,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499825941461198,
      "loss": 5.9108,
      "mean_token_accuracy": 0.12829016521573067,
      "num_tokens": 4947024.0,
      "step": 2250
    },
    {
      "entropy": 5.990304183959961,
      "epoch": 0.22379912663755458,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998245452701329,
      "loss": 5.8611,
      "mean_token_accuracy": 0.12755537852644921,
      "num_tokens": 4959281.0,
      "step": 2255
    },
    {
      "entropy": 6.069461011886597,
      "epoch": 0.2242953552997221,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998231435039036,
      "loss": 5.9047,
      "mean_token_accuracy": 0.12395669743418694,
      "num_tokens": 4970466.0,
      "step": 2260
    },
    {
      "entropy": 6.021377325057983,
      "epoch": 0.22479158396188964,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998217361625449,
      "loss": 5.8562,
      "mean_token_accuracy": 0.12923893854022026,
      "num_tokens": 4981135.0,
      "step": 2265
    },
    {
      "entropy": 6.048352479934692,
      "epoch": 0.22528781262405717,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998203232460915,
      "loss": 5.9167,
      "mean_token_accuracy": 0.1324473612010479,
      "num_tokens": 4991292.0,
      "step": 2270
    },
    {
      "entropy": 5.9737649917602536,
      "epoch": 0.2257840412862247,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998189047545785,
      "loss": 5.8825,
      "mean_token_accuracy": 0.13191651031374932,
      "num_tokens": 5002806.0,
      "step": 2275
    },
    {
      "entropy": 5.993149375915527,
      "epoch": 0.22628026994839223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998174806880411,
      "loss": 5.8179,
      "mean_token_accuracy": 0.13567918241024018,
      "num_tokens": 5013474.0,
      "step": 2280
    },
    {
      "entropy": 6.056148862838745,
      "epoch": 0.22677649861055973,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998160510465146,
      "loss": 5.8642,
      "mean_token_accuracy": 0.13143307566642762,
      "num_tokens": 5025151.0,
      "step": 2285
    },
    {
      "entropy": 5.92156720161438,
      "epoch": 0.22727272727272727,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998146158300345,
      "loss": 5.8638,
      "mean_token_accuracy": 0.13245996236801147,
      "num_tokens": 5037166.0,
      "step": 2290
    },
    {
      "entropy": 6.050130987167359,
      "epoch": 0.2277689559348948,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998131750386364,
      "loss": 5.9076,
      "mean_token_accuracy": 0.1297108441591263,
      "num_tokens": 5047452.0,
      "step": 2295
    },
    {
      "entropy": 6.028859281539917,
      "epoch": 0.22826518459706233,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998117286723559,
      "loss": 5.9272,
      "mean_token_accuracy": 0.11998689249157905,
      "num_tokens": 5058501.0,
      "step": 2300
    },
    {
      "entropy": 6.028175163269043,
      "epoch": 0.22876141325922986,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499810276731229,
      "loss": 5.9102,
      "mean_token_accuracy": 0.12801963984966278,
      "num_tokens": 5069716.0,
      "step": 2305
    },
    {
      "entropy": 6.123485803604126,
      "epoch": 0.2292576419213974,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998088192152918,
      "loss": 5.8893,
      "mean_token_accuracy": 0.1252819001674652,
      "num_tokens": 5078509.0,
      "step": 2310
    },
    {
      "entropy": 6.000077247619629,
      "epoch": 0.22975387058356492,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998073561245801,
      "loss": 5.9347,
      "mean_token_accuracy": 0.12345504015684128,
      "num_tokens": 5090355.0,
      "step": 2315
    },
    {
      "entropy": 6.051438713073731,
      "epoch": 0.23025009924573242,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998058874591305,
      "loss": 5.8242,
      "mean_token_accuracy": 0.1337508350610733,
      "num_tokens": 5100582.0,
      "step": 2320
    },
    {
      "entropy": 6.000838756561279,
      "epoch": 0.23074632790789995,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004998044132189793,
      "loss": 5.8887,
      "mean_token_accuracy": 0.12203488871455193,
      "num_tokens": 5112176.0,
      "step": 2325
    },
    {
      "entropy": 6.021178150177002,
      "epoch": 0.23124255657006748,
      "grad_norm": 1.28125,
      "learning_rate": 0.000499802933404163,
      "loss": 5.8335,
      "mean_token_accuracy": 0.1279439650475979,
      "num_tokens": 5122840.0,
      "step": 2330
    },
    {
      "entropy": 6.0098320007324215,
      "epoch": 0.231738785232235,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998014480147183,
      "loss": 5.8047,
      "mean_token_accuracy": 0.13865086659789086,
      "num_tokens": 5132926.0,
      "step": 2335
    },
    {
      "entropy": 5.934864521026611,
      "epoch": 0.23223501389440254,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997999570506823,
      "loss": 5.7806,
      "mean_token_accuracy": 0.13042209595441817,
      "num_tokens": 5143394.0,
      "step": 2340
    },
    {
      "entropy": 5.982978010177613,
      "epoch": 0.23273124255657007,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004997984605120917,
      "loss": 5.8147,
      "mean_token_accuracy": 0.12575569450855256,
      "num_tokens": 5155710.0,
      "step": 2345
    },
    {
      "entropy": 5.841717767715454,
      "epoch": 0.2332274712187376,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997969583989835,
      "loss": 5.751,
      "mean_token_accuracy": 0.12713968753814697,
      "num_tokens": 5167065.0,
      "step": 2350
    },
    {
      "entropy": 5.989433431625367,
      "epoch": 0.23372369988090513,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997954507113952,
      "loss": 5.8495,
      "mean_token_accuracy": 0.12923550233244896,
      "num_tokens": 5177535.0,
      "step": 2355
    },
    {
      "entropy": 6.037681818008423,
      "epoch": 0.23421992854307264,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004997939374493642,
      "loss": 5.7806,
      "mean_token_accuracy": 0.13312982767820358,
      "num_tokens": 5188324.0,
      "step": 2360
    },
    {
      "entropy": 5.901477909088134,
      "epoch": 0.23471615720524017,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000499792418612928,
      "loss": 5.86,
      "mean_token_accuracy": 0.1304488480091095,
      "num_tokens": 5199223.0,
      "step": 2365
    },
    {
      "entropy": 5.945672512054443,
      "epoch": 0.2352123858674077,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997908942021241,
      "loss": 5.8299,
      "mean_token_accuracy": 0.13325254768133163,
      "num_tokens": 5210080.0,
      "step": 2370
    },
    {
      "entropy": 6.099682092666626,
      "epoch": 0.23570861452957523,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997893642169904,
      "loss": 5.9558,
      "mean_token_accuracy": 0.12958926782011987,
      "num_tokens": 5220867.0,
      "step": 2375
    },
    {
      "entropy": 6.014858865737915,
      "epoch": 0.23620484319174276,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997878286575648,
      "loss": 5.7952,
      "mean_token_accuracy": 0.12957458645105363,
      "num_tokens": 5231451.0,
      "step": 2380
    },
    {
      "entropy": 5.938134288787841,
      "epoch": 0.2367010718539103,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997862875238855,
      "loss": 5.9034,
      "mean_token_accuracy": 0.12620984017848969,
      "num_tokens": 5244643.0,
      "step": 2385
    },
    {
      "entropy": 6.035072565078735,
      "epoch": 0.23719730051607782,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997847408159907,
      "loss": 5.9145,
      "mean_token_accuracy": 0.12331874147057534,
      "num_tokens": 5256279.0,
      "step": 2390
    },
    {
      "entropy": 6.0572350978851315,
      "epoch": 0.23769352917824532,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997831885339186,
      "loss": 5.8955,
      "mean_token_accuracy": 0.13024576678872107,
      "num_tokens": 5268498.0,
      "step": 2395
    },
    {
      "entropy": 5.939180183410644,
      "epoch": 0.23818975784041285,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997816306777077,
      "loss": 5.8567,
      "mean_token_accuracy": 0.13194234371185304,
      "num_tokens": 5280869.0,
      "step": 2400
    },
    {
      "entropy": 6.102878284454346,
      "epoch": 0.23868598650258038,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997800672473969,
      "loss": 5.9558,
      "mean_token_accuracy": 0.12405509725213051,
      "num_tokens": 5293082.0,
      "step": 2405
    },
    {
      "entropy": 5.935460233688355,
      "epoch": 0.2391822151647479,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997784982430247,
      "loss": 5.7407,
      "mean_token_accuracy": 0.14041346311569214,
      "num_tokens": 5304039.0,
      "step": 2410
    },
    {
      "entropy": 5.937913465499878,
      "epoch": 0.23967844382691544,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997769236646301,
      "loss": 5.8253,
      "mean_token_accuracy": 0.13146729916334152,
      "num_tokens": 5315016.0,
      "step": 2415
    },
    {
      "entropy": 6.090373134613037,
      "epoch": 0.24017467248908297,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997753435122522,
      "loss": 5.9481,
      "mean_token_accuracy": 0.12890507653355598,
      "num_tokens": 5326090.0,
      "step": 2420
    },
    {
      "entropy": 5.9364526748657225,
      "epoch": 0.2406709011512505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00049977375778593,
      "loss": 5.7381,
      "mean_token_accuracy": 0.13036757260560988,
      "num_tokens": 5336423.0,
      "step": 2425
    },
    {
      "entropy": 5.988410186767578,
      "epoch": 0.24116712981341804,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997721664857032,
      "loss": 5.9064,
      "mean_token_accuracy": 0.13083979934453965,
      "num_tokens": 5346982.0,
      "step": 2430
    },
    {
      "entropy": 5.9514707088470455,
      "epoch": 0.24166335847558554,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004997705696116108,
      "loss": 5.8041,
      "mean_token_accuracy": 0.13514370396733283,
      "num_tokens": 5359210.0,
      "step": 2435
    },
    {
      "entropy": 5.958845949172973,
      "epoch": 0.24215958713775307,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997689671636926,
      "loss": 5.8358,
      "mean_token_accuracy": 0.12565323561429978,
      "num_tokens": 5369891.0,
      "step": 2440
    },
    {
      "entropy": 5.962273597717285,
      "epoch": 0.2426558157999206,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004997673591419884,
      "loss": 5.8116,
      "mean_token_accuracy": 0.1305241070687771,
      "num_tokens": 5380850.0,
      "step": 2445
    },
    {
      "entropy": 5.970908737182617,
      "epoch": 0.24315204446208813,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997657455465381,
      "loss": 5.8724,
      "mean_token_accuracy": 0.13438860848546028,
      "num_tokens": 5391903.0,
      "step": 2450
    },
    {
      "entropy": 6.027713441848755,
      "epoch": 0.24364827312425566,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997641263773815,
      "loss": 5.9209,
      "mean_token_accuracy": 0.1283269263803959,
      "num_tokens": 5402517.0,
      "step": 2455
    },
    {
      "entropy": 5.9370378971099855,
      "epoch": 0.2441445017864232,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499762501634559,
      "loss": 5.8357,
      "mean_token_accuracy": 0.12951843738555907,
      "num_tokens": 5414110.0,
      "step": 2460
    },
    {
      "entropy": 6.02907428741455,
      "epoch": 0.24464073044859072,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997608713181108,
      "loss": 5.8403,
      "mean_token_accuracy": 0.13202603831887244,
      "num_tokens": 5424466.0,
      "step": 2465
    },
    {
      "entropy": 5.915480756759644,
      "epoch": 0.24513695911075822,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004997592354280772,
      "loss": 5.8526,
      "mean_token_accuracy": 0.1366075359284878,
      "num_tokens": 5434461.0,
      "step": 2470
    },
    {
      "entropy": 5.940920686721801,
      "epoch": 0.24563318777292575,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499757593964499,
      "loss": 5.8263,
      "mean_token_accuracy": 0.12434638366103172,
      "num_tokens": 5445921.0,
      "step": 2475
    },
    {
      "entropy": 5.955485916137695,
      "epoch": 0.24612941643509328,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997559469274166,
      "loss": 5.8127,
      "mean_token_accuracy": 0.12727661803364754,
      "num_tokens": 5456283.0,
      "step": 2480
    },
    {
      "entropy": 5.970708322525025,
      "epoch": 0.24662564509726081,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997542943168712,
      "loss": 5.8501,
      "mean_token_accuracy": 0.1340114578604698,
      "num_tokens": 5467616.0,
      "step": 2485
    },
    {
      "entropy": 5.974401187896729,
      "epoch": 0.24712187375942835,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997526361329035,
      "loss": 5.8048,
      "mean_token_accuracy": 0.13320422694087028,
      "num_tokens": 5478050.0,
      "step": 2490
    },
    {
      "entropy": 6.017988967895508,
      "epoch": 0.24761810242159588,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997509723755547,
      "loss": 5.8205,
      "mean_token_accuracy": 0.13056738600134848,
      "num_tokens": 5487185.0,
      "step": 2495
    },
    {
      "entropy": 5.945784282684326,
      "epoch": 0.2481143310837634,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997493030448662,
      "loss": 5.8411,
      "mean_token_accuracy": 0.1261320762336254,
      "num_tokens": 5498802.0,
      "step": 2500
    },
    {
      "entropy": 5.9388504981994625,
      "epoch": 0.24861055974593094,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997476281408793,
      "loss": 5.855,
      "mean_token_accuracy": 0.12759978100657463,
      "num_tokens": 5510285.0,
      "step": 2505
    },
    {
      "entropy": 6.00165810585022,
      "epoch": 0.24910678840809844,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997459476636354,
      "loss": 5.7335,
      "mean_token_accuracy": 0.13755400031805037,
      "num_tokens": 5520028.0,
      "step": 2510
    },
    {
      "entropy": 5.905316925048828,
      "epoch": 0.24960301707026597,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997442616131763,
      "loss": 5.85,
      "mean_token_accuracy": 0.12924634143710137,
      "num_tokens": 5529271.0,
      "step": 2515
    },
    {
      "entropy": 5.97735447883606,
      "epoch": 0.25009924573243353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997425699895438,
      "loss": 5.7753,
      "mean_token_accuracy": 0.12766662314534188,
      "num_tokens": 5539396.0,
      "step": 2520
    },
    {
      "entropy": 6.006867456436157,
      "epoch": 0.25059547439460106,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004997408727927798,
      "loss": 5.9196,
      "mean_token_accuracy": 0.13126150593161584,
      "num_tokens": 5551674.0,
      "step": 2525
    },
    {
      "entropy": 5.974583911895752,
      "epoch": 0.25109170305676853,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997391700229264,
      "loss": 5.8413,
      "mean_token_accuracy": 0.1326450914144516,
      "num_tokens": 5562749.0,
      "step": 2530
    },
    {
      "entropy": 5.960433483123779,
      "epoch": 0.25158793171893606,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499737461680026,
      "loss": 5.854,
      "mean_token_accuracy": 0.134577427059412,
      "num_tokens": 5573478.0,
      "step": 2535
    },
    {
      "entropy": 5.884474802017212,
      "epoch": 0.2520841603811036,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997357477641207,
      "loss": 5.8349,
      "mean_token_accuracy": 0.1368376225233078,
      "num_tokens": 5584546.0,
      "step": 2540
    },
    {
      "entropy": 5.904690790176391,
      "epoch": 0.2525803890432711,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997340282752532,
      "loss": 5.7177,
      "mean_token_accuracy": 0.13633103296160698,
      "num_tokens": 5594004.0,
      "step": 2545
    },
    {
      "entropy": 5.9720470905303955,
      "epoch": 0.25307661770543866,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499732303213466,
      "loss": 5.7943,
      "mean_token_accuracy": 0.125521769374609,
      "num_tokens": 5606095.0,
      "step": 2550
    },
    {
      "entropy": 5.999422073364258,
      "epoch": 0.2535728463676062,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004997305725788019,
      "loss": 5.813,
      "mean_token_accuracy": 0.12604894265532493,
      "num_tokens": 5615596.0,
      "step": 2555
    },
    {
      "entropy": 5.867432165145874,
      "epoch": 0.2540690750297737,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499728836371304,
      "loss": 5.771,
      "mean_token_accuracy": 0.13205990493297576,
      "num_tokens": 5626753.0,
      "step": 2560
    },
    {
      "entropy": 5.866491079330444,
      "epoch": 0.25456530369194125,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997270945910151,
      "loss": 5.7828,
      "mean_token_accuracy": 0.12868185043334962,
      "num_tokens": 5636135.0,
      "step": 2565
    },
    {
      "entropy": 6.046658754348755,
      "epoch": 0.2550615323541088,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997253472379785,
      "loss": 5.8504,
      "mean_token_accuracy": 0.1282991424202919,
      "num_tokens": 5647074.0,
      "step": 2570
    },
    {
      "entropy": 5.9704625606536865,
      "epoch": 0.2555577610162763,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004997235943122375,
      "loss": 5.8703,
      "mean_token_accuracy": 0.12940976694226264,
      "num_tokens": 5658600.0,
      "step": 2575
    },
    {
      "entropy": 5.924697065353394,
      "epoch": 0.25605398967844384,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997218358138356,
      "loss": 5.7866,
      "mean_token_accuracy": 0.13074240386486052,
      "num_tokens": 5668453.0,
      "step": 2580
    },
    {
      "entropy": 5.984611177444458,
      "epoch": 0.25655021834061137,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997200717428165,
      "loss": 5.8295,
      "mean_token_accuracy": 0.12661149725317955,
      "num_tokens": 5678855.0,
      "step": 2585
    },
    {
      "entropy": 5.916502237319946,
      "epoch": 0.2570464470027789,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004997183020992237,
      "loss": 5.8903,
      "mean_token_accuracy": 0.12172176912426949,
      "num_tokens": 5690063.0,
      "step": 2590
    },
    {
      "entropy": 5.936478853225708,
      "epoch": 0.25754267566494643,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997165268831014,
      "loss": 5.7896,
      "mean_token_accuracy": 0.13551302403211593,
      "num_tokens": 5701487.0,
      "step": 2595
    },
    {
      "entropy": 5.903074455261231,
      "epoch": 0.25803890432711396,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997147460944933,
      "loss": 5.7725,
      "mean_token_accuracy": 0.1349751755595207,
      "num_tokens": 5713034.0,
      "step": 2600
    },
    {
      "entropy": 6.014874839782715,
      "epoch": 0.25853513298928144,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004997129597334438,
      "loss": 5.8224,
      "mean_token_accuracy": 0.13061220347881317,
      "num_tokens": 5725469.0,
      "step": 2605
    },
    {
      "entropy": 5.862464809417725,
      "epoch": 0.25903136165144897,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004997111677999971,
      "loss": 5.812,
      "mean_token_accuracy": 0.12882769852876663,
      "num_tokens": 5737099.0,
      "step": 2610
    },
    {
      "entropy": 5.97240686416626,
      "epoch": 0.2595275903136165,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004997093702941976,
      "loss": 5.8214,
      "mean_token_accuracy": 0.12686606049537658,
      "num_tokens": 5748756.0,
      "step": 2615
    },
    {
      "entropy": 5.980477237701416,
      "epoch": 0.260023818975784,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049970756721609,
      "loss": 5.8099,
      "mean_token_accuracy": 0.12726252228021623,
      "num_tokens": 5759598.0,
      "step": 2620
    },
    {
      "entropy": 5.953548765182495,
      "epoch": 0.26052004763795156,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000499705758565719,
      "loss": 5.854,
      "mean_token_accuracy": 0.12930647283792496,
      "num_tokens": 5770397.0,
      "step": 2625
    },
    {
      "entropy": 6.005121231079102,
      "epoch": 0.2610162763001191,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997039443431292,
      "loss": 5.9263,
      "mean_token_accuracy": 0.12437670230865479,
      "num_tokens": 5780935.0,
      "step": 2630
    },
    {
      "entropy": 6.0060749530792235,
      "epoch": 0.2615125049622866,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997021245483659,
      "loss": 5.7808,
      "mean_token_accuracy": 0.13181432709097862,
      "num_tokens": 5791557.0,
      "step": 2635
    },
    {
      "entropy": 5.898467540740967,
      "epoch": 0.26200873362445415,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499700299181474,
      "loss": 5.8199,
      "mean_token_accuracy": 0.1286296509206295,
      "num_tokens": 5802233.0,
      "step": 2640
    },
    {
      "entropy": 5.9045013904571535,
      "epoch": 0.2625049622866217,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996984682424989,
      "loss": 5.7429,
      "mean_token_accuracy": 0.13722301945090293,
      "num_tokens": 5813424.0,
      "step": 2645
    },
    {
      "entropy": 5.888933420181274,
      "epoch": 0.2630011909487892,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000499696631731486,
      "loss": 5.7238,
      "mean_token_accuracy": 0.13307972252368927,
      "num_tokens": 5825652.0,
      "step": 2650
    },
    {
      "entropy": 5.953473615646362,
      "epoch": 0.26349741961095674,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996947896484809,
      "loss": 5.8126,
      "mean_token_accuracy": 0.13535162359476088,
      "num_tokens": 5836603.0,
      "step": 2655
    },
    {
      "entropy": 5.948557996749878,
      "epoch": 0.26399364827312427,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499692941993529,
      "loss": 5.8636,
      "mean_token_accuracy": 0.1336210548877716,
      "num_tokens": 5849365.0,
      "step": 2660
    },
    {
      "entropy": 5.975414085388183,
      "epoch": 0.2644898769352918,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996910887666765,
      "loss": 5.7797,
      "mean_token_accuracy": 0.1360326051712036,
      "num_tokens": 5860305.0,
      "step": 2665
    },
    {
      "entropy": 5.86191725730896,
      "epoch": 0.26498610559745933,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499689229967969,
      "loss": 5.7249,
      "mean_token_accuracy": 0.1364116244018078,
      "num_tokens": 5872272.0,
      "step": 2670
    },
    {
      "entropy": 5.931841802597046,
      "epoch": 0.26548233425962686,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996873655974529,
      "loss": 5.7536,
      "mean_token_accuracy": 0.13195567727088928,
      "num_tokens": 5884107.0,
      "step": 2675
    },
    {
      "entropy": 5.839572381973267,
      "epoch": 0.26597856292179434,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996854956551743,
      "loss": 5.7433,
      "mean_token_accuracy": 0.13461884558200837,
      "num_tokens": 5894060.0,
      "step": 2680
    },
    {
      "entropy": 5.867028617858887,
      "epoch": 0.26647479158396187,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996836201411794,
      "loss": 5.7661,
      "mean_token_accuracy": 0.13390844464302062,
      "num_tokens": 5904804.0,
      "step": 2685
    },
    {
      "entropy": 6.046043062210083,
      "epoch": 0.2669710202461294,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499681739055515,
      "loss": 5.8361,
      "mean_token_accuracy": 0.12845558002591134,
      "num_tokens": 5916496.0,
      "step": 2690
    },
    {
      "entropy": 5.9300178527832035,
      "epoch": 0.26746724890829693,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996798523982276,
      "loss": 5.7818,
      "mean_token_accuracy": 0.1285025343298912,
      "num_tokens": 5926920.0,
      "step": 2695
    },
    {
      "entropy": 5.965683031082153,
      "epoch": 0.26796347757046446,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499677960169364,
      "loss": 5.8865,
      "mean_token_accuracy": 0.12686679884791374,
      "num_tokens": 5940098.0,
      "step": 2700
    },
    {
      "entropy": 5.910781335830689,
      "epoch": 0.268459706232632,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996760623689711,
      "loss": 5.7374,
      "mean_token_accuracy": 0.13134501054883002,
      "num_tokens": 5949687.0,
      "step": 2705
    },
    {
      "entropy": 5.929884958267212,
      "epoch": 0.2689559348947995,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000499674158997096,
      "loss": 5.8442,
      "mean_token_accuracy": 0.12607498317956925,
      "num_tokens": 5960777.0,
      "step": 2710
    },
    {
      "entropy": 5.978472423553467,
      "epoch": 0.26945216355696705,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996722500537859,
      "loss": 5.7503,
      "mean_token_accuracy": 0.13114108070731162,
      "num_tokens": 5972298.0,
      "step": 2715
    },
    {
      "entropy": 5.902798414230347,
      "epoch": 0.2699483922191346,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996703355390881,
      "loss": 5.8696,
      "mean_token_accuracy": 0.12787102088332175,
      "num_tokens": 5985613.0,
      "step": 2720
    },
    {
      "entropy": 5.998373651504517,
      "epoch": 0.2704446208813021,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049966841545305,
      "loss": 5.9368,
      "mean_token_accuracy": 0.1289162516593933,
      "num_tokens": 5997530.0,
      "step": 2725
    },
    {
      "entropy": 5.979059410095215,
      "epoch": 0.27094084954346964,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996664897957195,
      "loss": 5.7906,
      "mean_token_accuracy": 0.12967565059661865,
      "num_tokens": 6008123.0,
      "step": 2730
    },
    {
      "entropy": 5.898594331741333,
      "epoch": 0.2714370782056372,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004996645585671441,
      "loss": 5.8159,
      "mean_token_accuracy": 0.13189680576324464,
      "num_tokens": 6020578.0,
      "step": 2735
    },
    {
      "entropy": 5.885920572280884,
      "epoch": 0.2719333068678047,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996626217673717,
      "loss": 5.7329,
      "mean_token_accuracy": 0.13771061301231385,
      "num_tokens": 6031625.0,
      "step": 2740
    },
    {
      "entropy": 5.859991264343262,
      "epoch": 0.27242953552997223,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004996606793964504,
      "loss": 5.771,
      "mean_token_accuracy": 0.13899061530828477,
      "num_tokens": 6041320.0,
      "step": 2745
    },
    {
      "entropy": 5.95008225440979,
      "epoch": 0.27292576419213976,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004996587314544284,
      "loss": 5.793,
      "mean_token_accuracy": 0.13212788924574853,
      "num_tokens": 6052627.0,
      "step": 2750
    },
    {
      "entropy": 5.905590677261353,
      "epoch": 0.27342199285430724,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996567779413539,
      "loss": 5.7442,
      "mean_token_accuracy": 0.13712375462055207,
      "num_tokens": 6063468.0,
      "step": 2755
    },
    {
      "entropy": 5.918176126480103,
      "epoch": 0.27391822151647477,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996548188572755,
      "loss": 5.7254,
      "mean_token_accuracy": 0.13772219493985177,
      "num_tokens": 6074208.0,
      "step": 2760
    },
    {
      "entropy": 5.93715763092041,
      "epoch": 0.2744144501786423,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004996528542022416,
      "loss": 5.7761,
      "mean_token_accuracy": 0.1317790240049362,
      "num_tokens": 6085068.0,
      "step": 2765
    },
    {
      "entropy": 5.8894490718841555,
      "epoch": 0.27491067884080983,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499650883976301,
      "loss": 5.7723,
      "mean_token_accuracy": 0.13286536261439325,
      "num_tokens": 6096445.0,
      "step": 2770
    },
    {
      "entropy": 5.8424560070037845,
      "epoch": 0.27540690750297736,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996489081795026,
      "loss": 5.6885,
      "mean_token_accuracy": 0.137538892775774,
      "num_tokens": 6106527.0,
      "step": 2775
    },
    {
      "entropy": 5.817946577072144,
      "epoch": 0.2759031361651449,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996469268118953,
      "loss": 5.7127,
      "mean_token_accuracy": 0.14144881442189217,
      "num_tokens": 6115923.0,
      "step": 2780
    },
    {
      "entropy": 5.838368940353393,
      "epoch": 0.2763993648273124,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996449398735283,
      "loss": 5.6099,
      "mean_token_accuracy": 0.14129377156496048,
      "num_tokens": 6125861.0,
      "step": 2785
    },
    {
      "entropy": 5.945985269546509,
      "epoch": 0.27689559348947995,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996429473644509,
      "loss": 5.8072,
      "mean_token_accuracy": 0.13425228521227836,
      "num_tokens": 6137189.0,
      "step": 2790
    },
    {
      "entropy": 5.8859100341796875,
      "epoch": 0.2773918221516475,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004996409492847124,
      "loss": 5.7305,
      "mean_token_accuracy": 0.13389328271150588,
      "num_tokens": 6147920.0,
      "step": 2795
    },
    {
      "entropy": 5.892881202697754,
      "epoch": 0.277888050813815,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004996389456343624,
      "loss": 5.7287,
      "mean_token_accuracy": 0.14077215641736984,
      "num_tokens": 6159371.0,
      "step": 2800
    },
    {
      "entropy": 6.015835094451904,
      "epoch": 0.27838427947598254,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004996369364134507,
      "loss": 5.9235,
      "mean_token_accuracy": 0.12242046073079109,
      "num_tokens": 6172679.0,
      "step": 2805
    },
    {
      "entropy": 5.894385576248169,
      "epoch": 0.2788805081381501,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499634921622027,
      "loss": 5.7718,
      "mean_token_accuracy": 0.12801498249173165,
      "num_tokens": 6184244.0,
      "step": 2810
    },
    {
      "entropy": 5.8594602108001705,
      "epoch": 0.2793767368003176,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996329012601413,
      "loss": 5.6501,
      "mean_token_accuracy": 0.1308717854321003,
      "num_tokens": 6194826.0,
      "step": 2815
    },
    {
      "entropy": 5.934268379211426,
      "epoch": 0.27987296546248513,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996308753278437,
      "loss": 5.7955,
      "mean_token_accuracy": 0.13045783266425132,
      "num_tokens": 6206808.0,
      "step": 2820
    },
    {
      "entropy": 5.903773975372315,
      "epoch": 0.28036919412465267,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996288438251843,
      "loss": 5.8036,
      "mean_token_accuracy": 0.1288069613277912,
      "num_tokens": 6217303.0,
      "step": 2825
    },
    {
      "entropy": 5.918454885482788,
      "epoch": 0.28086542278682014,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996268067522137,
      "loss": 5.7907,
      "mean_token_accuracy": 0.13076610937714578,
      "num_tokens": 6227965.0,
      "step": 2830
    },
    {
      "entropy": 6.045854377746582,
      "epoch": 0.28136165144898767,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996247641089823,
      "loss": 5.8534,
      "mean_token_accuracy": 0.12538014501333236,
      "num_tokens": 6239011.0,
      "step": 2835
    },
    {
      "entropy": 5.820587921142578,
      "epoch": 0.2818578801111552,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996227158955409,
      "loss": 5.7089,
      "mean_token_accuracy": 0.13381239399313927,
      "num_tokens": 6249084.0,
      "step": 2840
    },
    {
      "entropy": 5.854635334014892,
      "epoch": 0.28235410877332273,
      "grad_norm": 0.96875,
      "learning_rate": 0.00049962066211194,
      "loss": 5.7811,
      "mean_token_accuracy": 0.12858823388814927,
      "num_tokens": 6260706.0,
      "step": 2845
    },
    {
      "entropy": 5.910700273513794,
      "epoch": 0.28285033743549026,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996186027582308,
      "loss": 5.719,
      "mean_token_accuracy": 0.13755066841840743,
      "num_tokens": 6272848.0,
      "step": 2850
    },
    {
      "entropy": 5.831938123703003,
      "epoch": 0.2833465660976578,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996165378344642,
      "loss": 5.7198,
      "mean_token_accuracy": 0.1345575749874115,
      "num_tokens": 6283820.0,
      "step": 2855
    },
    {
      "entropy": 5.887689590454102,
      "epoch": 0.2838427947598253,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996144673406914,
      "loss": 5.8847,
      "mean_token_accuracy": 0.1203585371375084,
      "num_tokens": 6294193.0,
      "step": 2860
    },
    {
      "entropy": 5.846189975738525,
      "epoch": 0.28433902342199285,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996123912769639,
      "loss": 5.7111,
      "mean_token_accuracy": 0.14153243526816367,
      "num_tokens": 6304407.0,
      "step": 2865
    },
    {
      "entropy": 5.857138013839721,
      "epoch": 0.2848352520841604,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996103096433332,
      "loss": 5.7866,
      "mean_token_accuracy": 0.13395510241389275,
      "num_tokens": 6316311.0,
      "step": 2870
    },
    {
      "entropy": 5.944645738601684,
      "epoch": 0.2853314807463279,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996082224398506,
      "loss": 5.7471,
      "mean_token_accuracy": 0.13548968210816384,
      "num_tokens": 6327248.0,
      "step": 2875
    },
    {
      "entropy": 5.888983535766601,
      "epoch": 0.28582770940849545,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996061296665682,
      "loss": 5.7629,
      "mean_token_accuracy": 0.13373192697763442,
      "num_tokens": 6338494.0,
      "step": 2880
    },
    {
      "entropy": 5.86532621383667,
      "epoch": 0.286323938070663,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996040313235377,
      "loss": 5.6606,
      "mean_token_accuracy": 0.1401577278971672,
      "num_tokens": 6349038.0,
      "step": 2885
    },
    {
      "entropy": 5.888543939590454,
      "epoch": 0.2868201667328305,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996019274108112,
      "loss": 5.7737,
      "mean_token_accuracy": 0.12937842831015586,
      "num_tokens": 6360180.0,
      "step": 2890
    },
    {
      "entropy": 5.789392042160034,
      "epoch": 0.28731639539499804,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000499599817928441,
      "loss": 5.7128,
      "mean_token_accuracy": 0.13721143081784248,
      "num_tokens": 6370931.0,
      "step": 2895
    },
    {
      "entropy": 5.796161222457886,
      "epoch": 0.28781262405716557,
      "grad_norm": 0.875,
      "learning_rate": 0.0004995977028764792,
      "loss": 5.7034,
      "mean_token_accuracy": 0.14205339923501015,
      "num_tokens": 6381810.0,
      "step": 2900
    },
    {
      "entropy": 5.8964122295379635,
      "epoch": 0.28830885271933304,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995955822549783,
      "loss": 5.7075,
      "mean_token_accuracy": 0.1379415936768055,
      "num_tokens": 6392268.0,
      "step": 2905
    },
    {
      "entropy": 5.881337690353393,
      "epoch": 0.2888050813815006,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995934560639909,
      "loss": 5.7543,
      "mean_token_accuracy": 0.132367368042469,
      "num_tokens": 6403650.0,
      "step": 2910
    },
    {
      "entropy": 5.904406881332397,
      "epoch": 0.2893013100436681,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995913243035699,
      "loss": 5.7468,
      "mean_token_accuracy": 0.13186144679784775,
      "num_tokens": 6414672.0,
      "step": 2915
    },
    {
      "entropy": 5.880802202224731,
      "epoch": 0.28979753870583563,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995891869737678,
      "loss": 5.8088,
      "mean_token_accuracy": 0.1321220226585865,
      "num_tokens": 6424916.0,
      "step": 2920
    },
    {
      "entropy": 5.9205621719360355,
      "epoch": 0.29029376736800316,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995870440746379,
      "loss": 5.7555,
      "mean_token_accuracy": 0.13907312899827956,
      "num_tokens": 6435950.0,
      "step": 2925
    },
    {
      "entropy": 5.813934850692749,
      "epoch": 0.2907899960301707,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004995848956062333,
      "loss": 5.7223,
      "mean_token_accuracy": 0.1334433563053608,
      "num_tokens": 6446336.0,
      "step": 2930
    },
    {
      "entropy": 5.862525606155396,
      "epoch": 0.2912862246923382,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995827415686071,
      "loss": 5.7939,
      "mean_token_accuracy": 0.1313887841999531,
      "num_tokens": 6457655.0,
      "step": 2935
    },
    {
      "entropy": 5.933152437210083,
      "epoch": 0.29178245335450576,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499580581961813,
      "loss": 5.6738,
      "mean_token_accuracy": 0.1359466515481472,
      "num_tokens": 6469627.0,
      "step": 2940
    },
    {
      "entropy": 5.85411057472229,
      "epoch": 0.2922786820166733,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004995784167859043,
      "loss": 5.7708,
      "mean_token_accuracy": 0.1325305663049221,
      "num_tokens": 6480351.0,
      "step": 2945
    },
    {
      "entropy": 5.930560064315796,
      "epoch": 0.2927749106788408,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995762460409349,
      "loss": 5.7976,
      "mean_token_accuracy": 0.1347188226878643,
      "num_tokens": 6491651.0,
      "step": 2950
    },
    {
      "entropy": 5.843895006179809,
      "epoch": 0.29327113934100835,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004995740697269584,
      "loss": 5.6555,
      "mean_token_accuracy": 0.13501327559351922,
      "num_tokens": 6502531.0,
      "step": 2955
    },
    {
      "entropy": 5.776656246185302,
      "epoch": 0.2937673680031759,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995718878440289,
      "loss": 5.6508,
      "mean_token_accuracy": 0.13918386697769164,
      "num_tokens": 6511938.0,
      "step": 2960
    },
    {
      "entropy": 5.909338426589966,
      "epoch": 0.2942635966653434,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995697003922006,
      "loss": 5.8071,
      "mean_token_accuracy": 0.1281385228037834,
      "num_tokens": 6522775.0,
      "step": 2965
    },
    {
      "entropy": 5.889328289031982,
      "epoch": 0.29475982532751094,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995675073715274,
      "loss": 5.6384,
      "mean_token_accuracy": 0.1326303191483021,
      "num_tokens": 6534660.0,
      "step": 2970
    },
    {
      "entropy": 5.867802667617798,
      "epoch": 0.29525605398967847,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995653087820641,
      "loss": 5.7662,
      "mean_token_accuracy": 0.1363363467156887,
      "num_tokens": 6545673.0,
      "step": 2975
    },
    {
      "entropy": 5.872935438156128,
      "epoch": 0.29575228265184594,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499563104623865,
      "loss": 5.7364,
      "mean_token_accuracy": 0.13667785823345185,
      "num_tokens": 6556912.0,
      "step": 2980
    },
    {
      "entropy": 5.885501098632813,
      "epoch": 0.2962485113140135,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995608948969849,
      "loss": 5.8271,
      "mean_token_accuracy": 0.13097959756851196,
      "num_tokens": 6567532.0,
      "step": 2985
    },
    {
      "entropy": 5.90928168296814,
      "epoch": 0.296744739976181,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995586796014784,
      "loss": 5.796,
      "mean_token_accuracy": 0.13155989721417427,
      "num_tokens": 6577830.0,
      "step": 2990
    },
    {
      "entropy": 5.823900842666626,
      "epoch": 0.29724096863834853,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995564587374005,
      "loss": 5.6926,
      "mean_token_accuracy": 0.13435450717806816,
      "num_tokens": 6587254.0,
      "step": 2995
    },
    {
      "entropy": 5.902713966369629,
      "epoch": 0.29773719730051607,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004995542323048063,
      "loss": 5.8848,
      "mean_token_accuracy": 0.12699431478977202,
      "num_tokens": 6598294.0,
      "step": 3000
    },
    {
      "epoch": 0.29773719730051607,
      "eval_entropy": 5.728751434966033,
      "eval_loss": 5.813499450683594,
      "eval_mean_token_accuracy": 0.1359337329577737,
      "eval_num_tokens": 6598294.0,
      "eval_runtime": 20.8245,
      "eval_samples_per_second": 1616.798,
      "eval_steps_per_second": 202.118,
      "step": 3000
    },
    {
      "entropy": 5.949578332901001,
      "epoch": 0.2982334259626836,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499552000303751,
      "loss": 5.7558,
      "mean_token_accuracy": 0.13457091972231866,
      "num_tokens": 6608772.0,
      "step": 3005
    },
    {
      "entropy": 5.853357172012329,
      "epoch": 0.2987296546248511,
      "grad_norm": 0.90625,
      "learning_rate": 0.00049954976273429,
      "loss": 5.7668,
      "mean_token_accuracy": 0.13631939515471458,
      "num_tokens": 6619973.0,
      "step": 3010
    },
    {
      "entropy": 5.872531938552856,
      "epoch": 0.29922588328701866,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995475195964787,
      "loss": 5.782,
      "mean_token_accuracy": 0.13367341533303262,
      "num_tokens": 6631172.0,
      "step": 3015
    },
    {
      "entropy": 5.848301219940185,
      "epoch": 0.2997221119491862,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004995452708903727,
      "loss": 5.7763,
      "mean_token_accuracy": 0.13015636652708054,
      "num_tokens": 6641343.0,
      "step": 3020
    },
    {
      "entropy": 5.876894283294678,
      "epoch": 0.3002183406113537,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499543016616028,
      "loss": 5.7196,
      "mean_token_accuracy": 0.1363053061068058,
      "num_tokens": 6652892.0,
      "step": 3025
    },
    {
      "entropy": 5.898422145843506,
      "epoch": 0.30071456927352125,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004995407567735002,
      "loss": 5.7331,
      "mean_token_accuracy": 0.13731808662414552,
      "num_tokens": 6664177.0,
      "step": 3030
    },
    {
      "entropy": 5.860565662384033,
      "epoch": 0.3012107979356888,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995384913628456,
      "loss": 5.6982,
      "mean_token_accuracy": 0.13397607430815697,
      "num_tokens": 6674402.0,
      "step": 3035
    },
    {
      "entropy": 5.862356185913086,
      "epoch": 0.3017070265978563,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995362203841202,
      "loss": 5.8452,
      "mean_token_accuracy": 0.12625229880213737,
      "num_tokens": 6684886.0,
      "step": 3040
    },
    {
      "entropy": 5.876009559631347,
      "epoch": 0.30220325526002384,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995339438373804,
      "loss": 5.6665,
      "mean_token_accuracy": 0.1372012011706829,
      "num_tokens": 6694774.0,
      "step": 3045
    },
    {
      "entropy": 5.883480262756348,
      "epoch": 0.30269948392219137,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995316617226825,
      "loss": 5.7381,
      "mean_token_accuracy": 0.1354772590100765,
      "num_tokens": 6705346.0,
      "step": 3050
    },
    {
      "entropy": 5.802752208709717,
      "epoch": 0.30319571258435885,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995293740400833,
      "loss": 5.6781,
      "mean_token_accuracy": 0.14116692245006562,
      "num_tokens": 6715579.0,
      "step": 3055
    },
    {
      "entropy": 5.831467723846435,
      "epoch": 0.3036919412465264,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995270807896395,
      "loss": 5.6444,
      "mean_token_accuracy": 0.14167544841766358,
      "num_tokens": 6726382.0,
      "step": 3060
    },
    {
      "entropy": 5.8336011409759525,
      "epoch": 0.3041881699086939,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995247819714079,
      "loss": 5.7137,
      "mean_token_accuracy": 0.13432926014065744,
      "num_tokens": 6737245.0,
      "step": 3065
    },
    {
      "entropy": 5.889817237854004,
      "epoch": 0.30468439857086144,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995224775854454,
      "loss": 5.9247,
      "mean_token_accuracy": 0.12740089893341064,
      "num_tokens": 6749113.0,
      "step": 3070
    },
    {
      "entropy": 6.012298154830932,
      "epoch": 0.30518062723302897,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004995201676318093,
      "loss": 5.7805,
      "mean_token_accuracy": 0.13419591411948203,
      "num_tokens": 6761987.0,
      "step": 3075
    },
    {
      "entropy": 5.858676910400391,
      "epoch": 0.3056768558951965,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995178521105568,
      "loss": 5.7065,
      "mean_token_accuracy": 0.13019177466630935,
      "num_tokens": 6771279.0,
      "step": 3080
    },
    {
      "entropy": 5.801815176010132,
      "epoch": 0.30617308455736403,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004995155310217454,
      "loss": 5.6501,
      "mean_token_accuracy": 0.13885203823447229,
      "num_tokens": 6781038.0,
      "step": 3085
    },
    {
      "entropy": 5.84817328453064,
      "epoch": 0.30666931321953156,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995132043654327,
      "loss": 5.7956,
      "mean_token_accuracy": 0.13456297814846038,
      "num_tokens": 6792686.0,
      "step": 3090
    },
    {
      "entropy": 5.865736818313598,
      "epoch": 0.3071655418816991,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004995108721416763,
      "loss": 5.7511,
      "mean_token_accuracy": 0.1393015667796135,
      "num_tokens": 6804414.0,
      "step": 3095
    },
    {
      "entropy": 5.889206123352051,
      "epoch": 0.3076617705438666,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004995085343505339,
      "loss": 5.7185,
      "mean_token_accuracy": 0.13796254768967628,
      "num_tokens": 6816180.0,
      "step": 3100
    },
    {
      "entropy": 5.851957178115844,
      "epoch": 0.30815799920603415,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995061909920638,
      "loss": 5.7194,
      "mean_token_accuracy": 0.13511088341474534,
      "num_tokens": 6827114.0,
      "step": 3105
    },
    {
      "entropy": 5.898450088500977,
      "epoch": 0.3086542278682017,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995038420663238,
      "loss": 5.7754,
      "mean_token_accuracy": 0.13403737246990205,
      "num_tokens": 6836953.0,
      "step": 3110
    },
    {
      "entropy": 5.8968480110168455,
      "epoch": 0.3091504565303692,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995014875733724,
      "loss": 5.8208,
      "mean_token_accuracy": 0.12903424575924874,
      "num_tokens": 6847963.0,
      "step": 3115
    },
    {
      "entropy": 5.90019793510437,
      "epoch": 0.30964668519253674,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994991275132678,
      "loss": 5.8363,
      "mean_token_accuracy": 0.13068509995937347,
      "num_tokens": 6858264.0,
      "step": 3120
    },
    {
      "entropy": 5.890040588378906,
      "epoch": 0.31014291385470427,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994967618860685,
      "loss": 5.725,
      "mean_token_accuracy": 0.13812246397137642,
      "num_tokens": 6868861.0,
      "step": 3125
    },
    {
      "entropy": 5.879378128051758,
      "epoch": 0.31063914251687175,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994943906918333,
      "loss": 5.7576,
      "mean_token_accuracy": 0.1395530015230179,
      "num_tokens": 6879797.0,
      "step": 3130
    },
    {
      "entropy": 5.851549291610718,
      "epoch": 0.3111353711790393,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004994920139306211,
      "loss": 5.6931,
      "mean_token_accuracy": 0.1380985639989376,
      "num_tokens": 6891323.0,
      "step": 3135
    },
    {
      "entropy": 5.807861804962158,
      "epoch": 0.3116315998412068,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994896316024906,
      "loss": 5.7844,
      "mean_token_accuracy": 0.13645399808883668,
      "num_tokens": 6903345.0,
      "step": 3140
    },
    {
      "entropy": 5.850348901748657,
      "epoch": 0.31212782850337434,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499487243707501,
      "loss": 5.7002,
      "mean_token_accuracy": 0.1344092607498169,
      "num_tokens": 6914342.0,
      "step": 3145
    },
    {
      "entropy": 5.878314256668091,
      "epoch": 0.31262405716554187,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994848502457114,
      "loss": 5.7757,
      "mean_token_accuracy": 0.13398355096578599,
      "num_tokens": 6926691.0,
      "step": 3150
    },
    {
      "entropy": 5.797974109649658,
      "epoch": 0.3131202858277094,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994824512171812,
      "loss": 5.6434,
      "mean_token_accuracy": 0.1319235309958458,
      "num_tokens": 6937841.0,
      "step": 3155
    },
    {
      "entropy": 5.810340785980225,
      "epoch": 0.31361651448987693,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994800466219701,
      "loss": 5.7249,
      "mean_token_accuracy": 0.13448656126856803,
      "num_tokens": 6948601.0,
      "step": 3160
    },
    {
      "entropy": 5.866660022735596,
      "epoch": 0.31411274315204446,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994776364601374,
      "loss": 5.663,
      "mean_token_accuracy": 0.13846807479858397,
      "num_tokens": 6959775.0,
      "step": 3165
    },
    {
      "entropy": 5.769833517074585,
      "epoch": 0.314608971814212,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994752207317431,
      "loss": 5.6916,
      "mean_token_accuracy": 0.13938113674521446,
      "num_tokens": 6969202.0,
      "step": 3170
    },
    {
      "entropy": 5.793140506744384,
      "epoch": 0.3151052004763795,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499472799436847,
      "loss": 5.6714,
      "mean_token_accuracy": 0.1376807488501072,
      "num_tokens": 6978633.0,
      "step": 3175
    },
    {
      "entropy": 5.9285777568817135,
      "epoch": 0.31560142913854705,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994703725755092,
      "loss": 5.731,
      "mean_token_accuracy": 0.13278935626149177,
      "num_tokens": 6989638.0,
      "step": 3180
    },
    {
      "entropy": 5.860269212722779,
      "epoch": 0.3160976578007146,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994679401477899,
      "loss": 5.7104,
      "mean_token_accuracy": 0.13575214371085167,
      "num_tokens": 7001606.0,
      "step": 3185
    },
    {
      "entropy": 5.792490386962891,
      "epoch": 0.3165938864628821,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994655021537492,
      "loss": 5.6791,
      "mean_token_accuracy": 0.13449148833751678,
      "num_tokens": 7012915.0,
      "step": 3190
    },
    {
      "entropy": 5.800523090362549,
      "epoch": 0.31709011512504964,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994630585934478,
      "loss": 5.7451,
      "mean_token_accuracy": 0.13046567216515542,
      "num_tokens": 7023191.0,
      "step": 3195
    },
    {
      "entropy": 5.797707414627075,
      "epoch": 0.3175863437872172,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994606094669463,
      "loss": 5.6272,
      "mean_token_accuracy": 0.14198058173060418,
      "num_tokens": 7032506.0,
      "step": 3200
    },
    {
      "entropy": 5.8543891429901125,
      "epoch": 0.3180825724493847,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004994581547743053,
      "loss": 5.7768,
      "mean_token_accuracy": 0.12855609208345414,
      "num_tokens": 7046098.0,
      "step": 3205
    },
    {
      "entropy": 5.933334302902222,
      "epoch": 0.3185788011115522,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994556945155857,
      "loss": 5.7842,
      "mean_token_accuracy": 0.13946380093693733,
      "num_tokens": 7057434.0,
      "step": 3210
    },
    {
      "entropy": 5.781634664535522,
      "epoch": 0.3190750297737197,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994532286908486,
      "loss": 5.7094,
      "mean_token_accuracy": 0.1315760038793087,
      "num_tokens": 7067674.0,
      "step": 3215
    },
    {
      "entropy": 5.888899612426758,
      "epoch": 0.31957125843588724,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499450757300155,
      "loss": 5.7672,
      "mean_token_accuracy": 0.1315600000321865,
      "num_tokens": 7077898.0,
      "step": 3220
    },
    {
      "entropy": 5.922313070297241,
      "epoch": 0.32006748709805477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994482803435664,
      "loss": 5.7324,
      "mean_token_accuracy": 0.13344367668032647,
      "num_tokens": 7088566.0,
      "step": 3225
    },
    {
      "entropy": 5.8241376876831055,
      "epoch": 0.3205637157602223,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499445797821144,
      "loss": 5.7409,
      "mean_token_accuracy": 0.13400137275457383,
      "num_tokens": 7099533.0,
      "step": 3230
    },
    {
      "entropy": 5.858956956863404,
      "epoch": 0.32105994442238983,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994433097329494,
      "loss": 5.7326,
      "mean_token_accuracy": 0.13736375123262407,
      "num_tokens": 7112254.0,
      "step": 3235
    },
    {
      "entropy": 5.934496688842773,
      "epoch": 0.32155617308455736,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004994408160790444,
      "loss": 5.7663,
      "mean_token_accuracy": 0.1309788353741169,
      "num_tokens": 7124704.0,
      "step": 3240
    },
    {
      "entropy": 5.988117361068726,
      "epoch": 0.3220524017467249,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004994383168594907,
      "loss": 5.931,
      "mean_token_accuracy": 0.1246418759226799,
      "num_tokens": 7137741.0,
      "step": 3245
    },
    {
      "entropy": 5.875287818908691,
      "epoch": 0.3225486304088924,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994358120743506,
      "loss": 5.7035,
      "mean_token_accuracy": 0.1314437985420227,
      "num_tokens": 7147888.0,
      "step": 3250
    },
    {
      "entropy": 5.739679145812988,
      "epoch": 0.32304485907105995,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004994333017236858,
      "loss": 5.6589,
      "mean_token_accuracy": 0.1450750708580017,
      "num_tokens": 7158950.0,
      "step": 3255
    },
    {
      "entropy": 5.861381769180298,
      "epoch": 0.3235410877332275,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994307858075587,
      "loss": 5.8056,
      "mean_token_accuracy": 0.13047620952129363,
      "num_tokens": 7170156.0,
      "step": 3260
    },
    {
      "entropy": 5.854081630706787,
      "epoch": 0.324037316395395,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994282643260319,
      "loss": 5.6497,
      "mean_token_accuracy": 0.13866678774356841,
      "num_tokens": 7180576.0,
      "step": 3265
    },
    {
      "entropy": 5.776857566833496,
      "epoch": 0.32453354505756254,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004994257372791676,
      "loss": 5.7051,
      "mean_token_accuracy": 0.1358933262526989,
      "num_tokens": 7192380.0,
      "step": 3270
    },
    {
      "entropy": 5.895743036270142,
      "epoch": 0.3250297737197301,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004994232046670286,
      "loss": 5.7705,
      "mean_token_accuracy": 0.13492169752717018,
      "num_tokens": 7203898.0,
      "step": 3275
    },
    {
      "entropy": 5.9295378684997555,
      "epoch": 0.3255260023818976,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994206664896778,
      "loss": 5.7211,
      "mean_token_accuracy": 0.13575714454054832,
      "num_tokens": 7215241.0,
      "step": 3280
    },
    {
      "entropy": 5.788207483291626,
      "epoch": 0.3260222310440651,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994181227471779,
      "loss": 5.6898,
      "mean_token_accuracy": 0.13606705367565156,
      "num_tokens": 7226039.0,
      "step": 3285
    },
    {
      "entropy": 5.8000298023223875,
      "epoch": 0.3265184597062326,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004994155734395923,
      "loss": 5.7174,
      "mean_token_accuracy": 0.13560251593589784,
      "num_tokens": 7237148.0,
      "step": 3290
    },
    {
      "entropy": 5.844682121276856,
      "epoch": 0.32701468836840014,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499413018566984,
      "loss": 5.7074,
      "mean_token_accuracy": 0.13221757486462593,
      "num_tokens": 7249284.0,
      "step": 3295
    },
    {
      "entropy": 5.780368804931641,
      "epoch": 0.32751091703056767,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994104581294163,
      "loss": 5.7178,
      "mean_token_accuracy": 0.13415811136364936,
      "num_tokens": 7259967.0,
      "step": 3300
    },
    {
      "entropy": 5.746554756164551,
      "epoch": 0.3280071456927352,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004994078921269529,
      "loss": 5.6759,
      "mean_token_accuracy": 0.1414274677634239,
      "num_tokens": 7271145.0,
      "step": 3305
    },
    {
      "entropy": 5.904040813446045,
      "epoch": 0.32850337435490273,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994053205596573,
      "loss": 5.6909,
      "mean_token_accuracy": 0.13865874335169792,
      "num_tokens": 7281872.0,
      "step": 3310
    },
    {
      "entropy": 5.795816040039062,
      "epoch": 0.32899960301707026,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994027434275934,
      "loss": 5.7103,
      "mean_token_accuracy": 0.13829881325364113,
      "num_tokens": 7293330.0,
      "step": 3315
    },
    {
      "entropy": 5.911486101150513,
      "epoch": 0.3294958316792378,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004994001607308249,
      "loss": 5.7653,
      "mean_token_accuracy": 0.1293144330382347,
      "num_tokens": 7304187.0,
      "step": 3320
    },
    {
      "entropy": 5.881396007537842,
      "epoch": 0.3299920603414053,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499397572469416,
      "loss": 5.667,
      "mean_token_accuracy": 0.1366066187620163,
      "num_tokens": 7314756.0,
      "step": 3325
    },
    {
      "entropy": 5.746714353561401,
      "epoch": 0.33048828900357285,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004993949786434308,
      "loss": 5.7122,
      "mean_token_accuracy": 0.1379957266151905,
      "num_tokens": 7325711.0,
      "step": 3330
    },
    {
      "entropy": 5.778922271728516,
      "epoch": 0.3309845176657404,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993923792529337,
      "loss": 5.6361,
      "mean_token_accuracy": 0.14142280891537667,
      "num_tokens": 7335639.0,
      "step": 3335
    },
    {
      "entropy": 5.88486647605896,
      "epoch": 0.3314807463279079,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993897742979892,
      "loss": 5.7531,
      "mean_token_accuracy": 0.13650422394275666,
      "num_tokens": 7346621.0,
      "step": 3340
    },
    {
      "entropy": 5.86642370223999,
      "epoch": 0.33197697499007545,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004993871637786616,
      "loss": 5.7269,
      "mean_token_accuracy": 0.12956238090991973,
      "num_tokens": 7358950.0,
      "step": 3345
    },
    {
      "entropy": 5.761933898925781,
      "epoch": 0.332473203652243,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499384547695016,
      "loss": 5.5806,
      "mean_token_accuracy": 0.14225346446037293,
      "num_tokens": 7369629.0,
      "step": 3350
    },
    {
      "entropy": 5.821819877624511,
      "epoch": 0.3329694323144105,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499381926047117,
      "loss": 5.6669,
      "mean_token_accuracy": 0.14004866182804107,
      "num_tokens": 7379853.0,
      "step": 3355
    },
    {
      "entropy": 5.812605714797973,
      "epoch": 0.333465660976578,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00049937929883503,
      "loss": 5.7162,
      "mean_token_accuracy": 0.13662909269332885,
      "num_tokens": 7391323.0,
      "step": 3360
    },
    {
      "entropy": 5.839717102050781,
      "epoch": 0.3339618896387455,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004993766660588196,
      "loss": 5.6707,
      "mean_token_accuracy": 0.1323630690574646,
      "num_tokens": 7402201.0,
      "step": 3365
    },
    {
      "entropy": 5.884400796890259,
      "epoch": 0.33445811830091304,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993740277185515,
      "loss": 5.9218,
      "mean_token_accuracy": 0.12637710124254226,
      "num_tokens": 7413783.0,
      "step": 3370
    },
    {
      "entropy": 5.816495943069458,
      "epoch": 0.3349543469630806,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499371383814291,
      "loss": 5.6916,
      "mean_token_accuracy": 0.1357066087424755,
      "num_tokens": 7425909.0,
      "step": 3375
    },
    {
      "entropy": 5.770580053329468,
      "epoch": 0.3354505756252481,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993687343461036,
      "loss": 5.5844,
      "mean_token_accuracy": 0.14517004415392876,
      "num_tokens": 7436346.0,
      "step": 3380
    },
    {
      "entropy": 5.808633470535279,
      "epoch": 0.33594680428741563,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004993660793140551,
      "loss": 5.6471,
      "mean_token_accuracy": 0.14266916885972022,
      "num_tokens": 7446596.0,
      "step": 3385
    },
    {
      "entropy": 5.769165945053101,
      "epoch": 0.33644303294958317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993634187182113,
      "loss": 5.7441,
      "mean_token_accuracy": 0.13104177489876748,
      "num_tokens": 7457697.0,
      "step": 3390
    },
    {
      "entropy": 5.9399810314178465,
      "epoch": 0.3369392616117507,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993607525586381,
      "loss": 5.8457,
      "mean_token_accuracy": 0.13390295282006265,
      "num_tokens": 7468541.0,
      "step": 3395
    },
    {
      "entropy": 5.8909464359283445,
      "epoch": 0.3374354902739182,
      "grad_norm": 0.875,
      "learning_rate": 0.0004993580808354018,
      "loss": 5.726,
      "mean_token_accuracy": 0.1409744441509247,
      "num_tokens": 7479781.0,
      "step": 3400
    },
    {
      "entropy": 5.798458576202393,
      "epoch": 0.33793171893608576,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993554035485685,
      "loss": 5.6354,
      "mean_token_accuracy": 0.13420383185148238,
      "num_tokens": 7490722.0,
      "step": 3405
    },
    {
      "entropy": 5.815723514556884,
      "epoch": 0.3384279475982533,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993527206982047,
      "loss": 5.7456,
      "mean_token_accuracy": 0.1327146239578724,
      "num_tokens": 7502293.0,
      "step": 3410
    },
    {
      "entropy": 5.8476334571838375,
      "epoch": 0.3389241762604208,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004993500322843767,
      "loss": 5.6813,
      "mean_token_accuracy": 0.14118870049715043,
      "num_tokens": 7513846.0,
      "step": 3415
    },
    {
      "entropy": 5.772971153259277,
      "epoch": 0.33942040492258835,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993473383071514,
      "loss": 5.6888,
      "mean_token_accuracy": 0.13772222474217416,
      "num_tokens": 7524062.0,
      "step": 3420
    },
    {
      "entropy": 5.8234320163726805,
      "epoch": 0.3399166335847559,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993446387665956,
      "loss": 5.7599,
      "mean_token_accuracy": 0.13640008345246316,
      "num_tokens": 7535533.0,
      "step": 3425
    },
    {
      "entropy": 5.834895324707031,
      "epoch": 0.3404128622469234,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993419336627761,
      "loss": 5.6559,
      "mean_token_accuracy": 0.13569505959749223,
      "num_tokens": 7545951.0,
      "step": 3430
    },
    {
      "entropy": 5.771151876449585,
      "epoch": 0.3409090909090909,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004993392229957602,
      "loss": 5.596,
      "mean_token_accuracy": 0.13585764318704605,
      "num_tokens": 7557320.0,
      "step": 3435
    },
    {
      "entropy": 5.7666747093200685,
      "epoch": 0.3414053195712584,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993365067656148,
      "loss": 5.708,
      "mean_token_accuracy": 0.13677700236439705,
      "num_tokens": 7568078.0,
      "step": 3440
    },
    {
      "entropy": 5.792825222015381,
      "epoch": 0.34190154823342594,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004993337849724076,
      "loss": 5.6956,
      "mean_token_accuracy": 0.13778189420700074,
      "num_tokens": 7578940.0,
      "step": 3445
    },
    {
      "entropy": 5.818231916427612,
      "epoch": 0.3423977768955935,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993310576162059,
      "loss": 5.7024,
      "mean_token_accuracy": 0.1383884370326996,
      "num_tokens": 7589864.0,
      "step": 3450
    },
    {
      "entropy": 5.852412557601928,
      "epoch": 0.342894005557761,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993283246970772,
      "loss": 5.6502,
      "mean_token_accuracy": 0.13423066288232804,
      "num_tokens": 7599795.0,
      "step": 3455
    },
    {
      "entropy": 5.793929576873779,
      "epoch": 0.34339023421992854,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993255862150896,
      "loss": 5.6382,
      "mean_token_accuracy": 0.1384445235133171,
      "num_tokens": 7610056.0,
      "step": 3460
    },
    {
      "entropy": 5.756991195678711,
      "epoch": 0.34388646288209607,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993228421703108,
      "loss": 5.6375,
      "mean_token_accuracy": 0.13536739200353623,
      "num_tokens": 7620218.0,
      "step": 3465
    },
    {
      "entropy": 5.7849561214447025,
      "epoch": 0.3443826915442636,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004993200925628088,
      "loss": 5.7457,
      "mean_token_accuracy": 0.13547196015715599,
      "num_tokens": 7632781.0,
      "step": 3470
    },
    {
      "entropy": 5.846343946456909,
      "epoch": 0.3448789202064311,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000499317337392652,
      "loss": 5.7067,
      "mean_token_accuracy": 0.13641866371035577,
      "num_tokens": 7643845.0,
      "step": 3475
    },
    {
      "entropy": 5.814862680435181,
      "epoch": 0.34537514886859866,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993145766599085,
      "loss": 5.7323,
      "mean_token_accuracy": 0.13265105634927749,
      "num_tokens": 7654583.0,
      "step": 3480
    },
    {
      "entropy": 5.779863452911377,
      "epoch": 0.3458713775307662,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004993118103646469,
      "loss": 5.581,
      "mean_token_accuracy": 0.14222516641020774,
      "num_tokens": 7665714.0,
      "step": 3485
    },
    {
      "entropy": 5.807524061203003,
      "epoch": 0.3463676061929337,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004993090385069357,
      "loss": 5.6426,
      "mean_token_accuracy": 0.1403677485883236,
      "num_tokens": 7676011.0,
      "step": 3490
    },
    {
      "entropy": 5.743056392669677,
      "epoch": 0.34686383485510125,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004993062610868437,
      "loss": 5.6994,
      "mean_token_accuracy": 0.13592455089092254,
      "num_tokens": 7686927.0,
      "step": 3495
    },
    {
      "entropy": 5.7684183597564695,
      "epoch": 0.3473600635172688,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004993034781044398,
      "loss": 5.6349,
      "mean_token_accuracy": 0.13586694225668908,
      "num_tokens": 7698419.0,
      "step": 3500
    },
    {
      "entropy": 5.849251174926758,
      "epoch": 0.3478562921794363,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499300689559793,
      "loss": 5.627,
      "mean_token_accuracy": 0.13978302255272865,
      "num_tokens": 7709695.0,
      "step": 3505
    },
    {
      "entropy": 5.78063006401062,
      "epoch": 0.3483525208416038,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004992978954529724,
      "loss": 5.6018,
      "mean_token_accuracy": 0.14354961216449738,
      "num_tokens": 7721309.0,
      "step": 3510
    },
    {
      "entropy": 5.775903558731079,
      "epoch": 0.3488487495037713,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992950957840473,
      "loss": 5.6792,
      "mean_token_accuracy": 0.13921934366226196,
      "num_tokens": 7732074.0,
      "step": 3515
    },
    {
      "entropy": 5.859705686569214,
      "epoch": 0.34934497816593885,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499292290553087,
      "loss": 5.7478,
      "mean_token_accuracy": 0.13691921010613442,
      "num_tokens": 7742869.0,
      "step": 3520
    },
    {
      "entropy": 5.745819091796875,
      "epoch": 0.3498412068281064,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004992894797601614,
      "loss": 5.6456,
      "mean_token_accuracy": 0.13721488118171693,
      "num_tokens": 7754895.0,
      "step": 3525
    },
    {
      "entropy": 5.811453056335449,
      "epoch": 0.3503374354902739,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049928666340534,
      "loss": 5.613,
      "mean_token_accuracy": 0.1413191042840481,
      "num_tokens": 7765262.0,
      "step": 3530
    },
    {
      "entropy": 5.722553396224976,
      "epoch": 0.35083366415244144,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004992838414886925,
      "loss": 5.6921,
      "mean_token_accuracy": 0.13405056223273276,
      "num_tokens": 7777348.0,
      "step": 3535
    },
    {
      "entropy": 5.832651519775391,
      "epoch": 0.35132989281460897,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992810140102891,
      "loss": 5.6802,
      "mean_token_accuracy": 0.136273655295372,
      "num_tokens": 7788161.0,
      "step": 3540
    },
    {
      "entropy": 5.782558441162109,
      "epoch": 0.3518261214767765,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004992781809701999,
      "loss": 5.6726,
      "mean_token_accuracy": 0.1378345899283886,
      "num_tokens": 7798629.0,
      "step": 3545
    },
    {
      "entropy": 5.735908031463623,
      "epoch": 0.35232235013894403,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992753423684951,
      "loss": 5.6783,
      "mean_token_accuracy": 0.137686438113451,
      "num_tokens": 7810148.0,
      "step": 3550
    },
    {
      "entropy": 5.815752363204956,
      "epoch": 0.35281857880111156,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992724982052449,
      "loss": 5.6681,
      "mean_token_accuracy": 0.13577194958925248,
      "num_tokens": 7821667.0,
      "step": 3555
    },
    {
      "entropy": 5.76047043800354,
      "epoch": 0.3533148074632791,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00049926964848052,
      "loss": 5.653,
      "mean_token_accuracy": 0.14100263491272927,
      "num_tokens": 7832184.0,
      "step": 3560
    },
    {
      "entropy": 5.8020179271698,
      "epoch": 0.3538110361254466,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992667931943913,
      "loss": 5.6656,
      "mean_token_accuracy": 0.13793855458498,
      "num_tokens": 7842983.0,
      "step": 3565
    },
    {
      "entropy": 5.791328716278076,
      "epoch": 0.35430726478761415,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004992639323469293,
      "loss": 5.6032,
      "mean_token_accuracy": 0.14528444707393645,
      "num_tokens": 7854783.0,
      "step": 3570
    },
    {
      "entropy": 5.734807443618775,
      "epoch": 0.3548034934497817,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499261065938205,
      "loss": 5.6097,
      "mean_token_accuracy": 0.14882165640592576,
      "num_tokens": 7864684.0,
      "step": 3575
    },
    {
      "entropy": 5.812001514434814,
      "epoch": 0.3552997221119492,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992581939682895,
      "loss": 5.774,
      "mean_token_accuracy": 0.1278185188770294,
      "num_tokens": 7876846.0,
      "step": 3580
    },
    {
      "entropy": 5.900391387939453,
      "epoch": 0.3557959507741167,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499255316437254,
      "loss": 5.64,
      "mean_token_accuracy": 0.13711518049240112,
      "num_tokens": 7886826.0,
      "step": 3585
    },
    {
      "entropy": 5.703480863571167,
      "epoch": 0.3562921794362842,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992524333451699,
      "loss": 5.7114,
      "mean_token_accuracy": 0.13370427638292312,
      "num_tokens": 7898941.0,
      "step": 3590
    },
    {
      "entropy": 5.773089456558227,
      "epoch": 0.35678840809845175,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992495446921087,
      "loss": 5.616,
      "mean_token_accuracy": 0.1364353835582733,
      "num_tokens": 7909916.0,
      "step": 3595
    },
    {
      "entropy": 5.7855347156524655,
      "epoch": 0.3572846367606193,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992466504781419,
      "loss": 5.6485,
      "mean_token_accuracy": 0.13997598960995675,
      "num_tokens": 7922137.0,
      "step": 3600
    },
    {
      "entropy": 5.682146310806274,
      "epoch": 0.3577808654227868,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004992437507033415,
      "loss": 5.5599,
      "mean_token_accuracy": 0.14679991602897643,
      "num_tokens": 7932940.0,
      "step": 3605
    },
    {
      "entropy": 5.837021684646606,
      "epoch": 0.35827709408495434,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992408453677792,
      "loss": 5.7332,
      "mean_token_accuracy": 0.13032081127166747,
      "num_tokens": 7943735.0,
      "step": 3610
    },
    {
      "entropy": 5.760317420959472,
      "epoch": 0.35877332274712187,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992379344715273,
      "loss": 5.6161,
      "mean_token_accuracy": 0.13913318663835525,
      "num_tokens": 7955323.0,
      "step": 3615
    },
    {
      "entropy": 5.6868006706237795,
      "epoch": 0.3592695514092894,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992350180146577,
      "loss": 5.5981,
      "mean_token_accuracy": 0.14091636314988137,
      "num_tokens": 7966073.0,
      "step": 3620
    },
    {
      "entropy": 5.820367622375488,
      "epoch": 0.35976578007145693,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992320959972428,
      "loss": 5.7517,
      "mean_token_accuracy": 0.13469664603471757,
      "num_tokens": 7977027.0,
      "step": 3625
    },
    {
      "entropy": 5.770542049407959,
      "epoch": 0.36026200873362446,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004992291684193551,
      "loss": 5.6092,
      "mean_token_accuracy": 0.14524339586496354,
      "num_tokens": 7987368.0,
      "step": 3630
    },
    {
      "entropy": 5.802702474594116,
      "epoch": 0.360758237395792,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992262352810672,
      "loss": 5.6917,
      "mean_token_accuracy": 0.13706899285316468,
      "num_tokens": 8000060.0,
      "step": 3635
    },
    {
      "entropy": 5.845965242385864,
      "epoch": 0.3612544660579595,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992232965824518,
      "loss": 5.7226,
      "mean_token_accuracy": 0.13262399807572364,
      "num_tokens": 8011388.0,
      "step": 3640
    },
    {
      "entropy": 5.746702432632446,
      "epoch": 0.36175069472012705,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499220352323582,
      "loss": 5.605,
      "mean_token_accuracy": 0.14428586438298224,
      "num_tokens": 8021682.0,
      "step": 3645
    },
    {
      "entropy": 5.750078248977661,
      "epoch": 0.3622469233822946,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992174025045304,
      "loss": 5.6234,
      "mean_token_accuracy": 0.13938501179218293,
      "num_tokens": 8032799.0,
      "step": 3650
    },
    {
      "entropy": 5.704130172729492,
      "epoch": 0.3627431520444621,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992144471253704,
      "loss": 5.5055,
      "mean_token_accuracy": 0.14247506186366082,
      "num_tokens": 8042742.0,
      "step": 3655
    },
    {
      "entropy": 5.786503791809082,
      "epoch": 0.3632393807066296,
      "grad_norm": 0.875,
      "learning_rate": 0.0004992114861861753,
      "loss": 5.6827,
      "mean_token_accuracy": 0.13944674730300904,
      "num_tokens": 8054068.0,
      "step": 3660
    },
    {
      "entropy": 5.790756273269653,
      "epoch": 0.3637356093687971,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004992085196870184,
      "loss": 5.6695,
      "mean_token_accuracy": 0.13868657797574996,
      "num_tokens": 8065286.0,
      "step": 3665
    },
    {
      "entropy": 5.766940307617188,
      "epoch": 0.36423183803096465,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004992055476279734,
      "loss": 5.7167,
      "mean_token_accuracy": 0.12711743265390396,
      "num_tokens": 8077122.0,
      "step": 3670
    },
    {
      "entropy": 5.846836996078491,
      "epoch": 0.3647280666931322,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992025700091139,
      "loss": 5.66,
      "mean_token_accuracy": 0.1416809819638729,
      "num_tokens": 8087036.0,
      "step": 3675
    },
    {
      "entropy": 5.781601667404175,
      "epoch": 0.3652242953552997,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004991995868305138,
      "loss": 5.6973,
      "mean_token_accuracy": 0.1388478934764862,
      "num_tokens": 8098517.0,
      "step": 3680
    },
    {
      "entropy": 5.680510234832764,
      "epoch": 0.36572052401746724,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991965980922471,
      "loss": 5.5717,
      "mean_token_accuracy": 0.14402113407850264,
      "num_tokens": 8107475.0,
      "step": 3685
    },
    {
      "entropy": 5.778713607788086,
      "epoch": 0.36621675267963477,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004991936037943878,
      "loss": 5.6709,
      "mean_token_accuracy": 0.13664408326148986,
      "num_tokens": 8118529.0,
      "step": 3690
    },
    {
      "entropy": 5.82429313659668,
      "epoch": 0.3667129813418023,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991906039370104,
      "loss": 5.5781,
      "mean_token_accuracy": 0.14057729691267012,
      "num_tokens": 8128951.0,
      "step": 3695
    },
    {
      "entropy": 5.6867591381073,
      "epoch": 0.36720921000396983,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004991875985201889,
      "loss": 5.5914,
      "mean_token_accuracy": 0.14267491921782494,
      "num_tokens": 8139385.0,
      "step": 3700
    },
    {
      "entropy": 5.760690116882325,
      "epoch": 0.36770543866613736,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499184587543998,
      "loss": 5.6417,
      "mean_token_accuracy": 0.14163167104125024,
      "num_tokens": 8149376.0,
      "step": 3705
    },
    {
      "entropy": 5.830694532394409,
      "epoch": 0.3682016673283049,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991815710085126,
      "loss": 5.6757,
      "mean_token_accuracy": 0.13829845562577248,
      "num_tokens": 8160687.0,
      "step": 3710
    },
    {
      "entropy": 5.783746433258057,
      "epoch": 0.3686978959904724,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004991785489138072,
      "loss": 5.673,
      "mean_token_accuracy": 0.13856422975659372,
      "num_tokens": 8171762.0,
      "step": 3715
    },
    {
      "entropy": 5.668252658843994,
      "epoch": 0.36919412465263995,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991755212599569,
      "loss": 5.4415,
      "mean_token_accuracy": 0.14850418344140054,
      "num_tokens": 8181643.0,
      "step": 3720
    },
    {
      "entropy": 5.653321266174316,
      "epoch": 0.3696903533148075,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991724880470366,
      "loss": 5.6961,
      "mean_token_accuracy": 0.14297981560230255,
      "num_tokens": 8192692.0,
      "step": 3725
    },
    {
      "entropy": 5.8245847702026365,
      "epoch": 0.370186581976975,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004991694492751217,
      "loss": 5.6274,
      "mean_token_accuracy": 0.1364683873951435,
      "num_tokens": 8202941.0,
      "step": 3730
    },
    {
      "entropy": 5.747600936889649,
      "epoch": 0.3706828106391425,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004991664049442876,
      "loss": 5.6263,
      "mean_token_accuracy": 0.14124569520354271,
      "num_tokens": 8214206.0,
      "step": 3735
    },
    {
      "entropy": 5.694499111175537,
      "epoch": 0.37117903930131,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004991633550546097,
      "loss": 5.4604,
      "mean_token_accuracy": 0.1512777641415596,
      "num_tokens": 8223672.0,
      "step": 3740
    },
    {
      "entropy": 5.720300483703613,
      "epoch": 0.37167526796347755,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004991602996061635,
      "loss": 5.6137,
      "mean_token_accuracy": 0.14432198852300643,
      "num_tokens": 8234189.0,
      "step": 3745
    },
    {
      "entropy": 5.8249866485595705,
      "epoch": 0.3721714966256451,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004991572385990249,
      "loss": 5.6509,
      "mean_token_accuracy": 0.14039039388298988,
      "num_tokens": 8245382.0,
      "step": 3750
    },
    {
      "entropy": 5.78410153388977,
      "epoch": 0.3726677252878126,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991541720332698,
      "loss": 5.6006,
      "mean_token_accuracy": 0.14047230705618857,
      "num_tokens": 8254530.0,
      "step": 3755
    },
    {
      "entropy": 5.685094213485717,
      "epoch": 0.37316395394998014,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991510999089743,
      "loss": 5.6452,
      "mean_token_accuracy": 0.1451885238289833,
      "num_tokens": 8264030.0,
      "step": 3760
    },
    {
      "entropy": 5.694255018234253,
      "epoch": 0.3736601826121477,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004991480222262145,
      "loss": 5.6106,
      "mean_token_accuracy": 0.14497761949896812,
      "num_tokens": 8273537.0,
      "step": 3765
    },
    {
      "entropy": 5.816177749633789,
      "epoch": 0.3741564112743152,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004991449389850668,
      "loss": 5.6572,
      "mean_token_accuracy": 0.13783644288778304,
      "num_tokens": 8286123.0,
      "step": 3770
    },
    {
      "entropy": 5.77542405128479,
      "epoch": 0.37465263993648273,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004991418501856076,
      "loss": 5.6482,
      "mean_token_accuracy": 0.13695365637540818,
      "num_tokens": 8298115.0,
      "step": 3775
    },
    {
      "entropy": 5.780150699615478,
      "epoch": 0.37514886859865026,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991387558279134,
      "loss": 5.6735,
      "mean_token_accuracy": 0.13566352427005768,
      "num_tokens": 8309376.0,
      "step": 3780
    },
    {
      "entropy": 5.761322498321533,
      "epoch": 0.3756450972608178,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991356559120612,
      "loss": 5.6279,
      "mean_token_accuracy": 0.13963163048028945,
      "num_tokens": 8319272.0,
      "step": 3785
    },
    {
      "entropy": 5.761824226379394,
      "epoch": 0.3761413259229853,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991325504381275,
      "loss": 5.629,
      "mean_token_accuracy": 0.13797441720962525,
      "num_tokens": 8330042.0,
      "step": 3790
    },
    {
      "entropy": 5.743603467941284,
      "epoch": 0.37663755458515286,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004991294394061896,
      "loss": 5.596,
      "mean_token_accuracy": 0.14118768870830536,
      "num_tokens": 8340394.0,
      "step": 3795
    },
    {
      "entropy": 5.70108060836792,
      "epoch": 0.3771337832473204,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004991263228163246,
      "loss": 5.6309,
      "mean_token_accuracy": 0.1392921596765518,
      "num_tokens": 8351103.0,
      "step": 3800
    },
    {
      "entropy": 5.783044767379761,
      "epoch": 0.3776300119094879,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004991232006686096,
      "loss": 5.6576,
      "mean_token_accuracy": 0.13801991567015648,
      "num_tokens": 8363215.0,
      "step": 3805
    },
    {
      "entropy": 5.8357092380523685,
      "epoch": 0.3781262405716554,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991200729631223,
      "loss": 5.7,
      "mean_token_accuracy": 0.13135057240724562,
      "num_tokens": 8374602.0,
      "step": 3810
    },
    {
      "entropy": 5.7686113834381105,
      "epoch": 0.3786224692338229,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991169396999401,
      "loss": 5.6115,
      "mean_token_accuracy": 0.13531650006771087,
      "num_tokens": 8386470.0,
      "step": 3815
    },
    {
      "entropy": 5.760716915130615,
      "epoch": 0.37911869789599045,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991138008791408,
      "loss": 5.6257,
      "mean_token_accuracy": 0.14137942716479301,
      "num_tokens": 8396284.0,
      "step": 3820
    },
    {
      "entropy": 5.784114170074463,
      "epoch": 0.379614926558158,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004991106565008021,
      "loss": 5.6669,
      "mean_token_accuracy": 0.13694209456443787,
      "num_tokens": 8407856.0,
      "step": 3825
    },
    {
      "entropy": 5.8008862972259525,
      "epoch": 0.3801111552203255,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004991075065650019,
      "loss": 5.5639,
      "mean_token_accuracy": 0.14441588819026946,
      "num_tokens": 8419560.0,
      "step": 3830
    },
    {
      "entropy": 5.788234567642212,
      "epoch": 0.38060738388249304,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991043510718186,
      "loss": 5.5603,
      "mean_token_accuracy": 0.14501854926347732,
      "num_tokens": 8430594.0,
      "step": 3835
    },
    {
      "entropy": 5.681872177124023,
      "epoch": 0.3811036125446606,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004991011900213303,
      "loss": 5.5325,
      "mean_token_accuracy": 0.14425235539674758,
      "num_tokens": 8440549.0,
      "step": 3840
    },
    {
      "entropy": 5.670712041854858,
      "epoch": 0.3815998412068281,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004990980234136152,
      "loss": 5.579,
      "mean_token_accuracy": 0.1425841838121414,
      "num_tokens": 8451553.0,
      "step": 3845
    },
    {
      "entropy": 5.710864925384522,
      "epoch": 0.38209606986899564,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004990948512487522,
      "loss": 5.5829,
      "mean_token_accuracy": 0.14589926972985268,
      "num_tokens": 8462899.0,
      "step": 3850
    },
    {
      "entropy": 5.752317094802857,
      "epoch": 0.38259229853116317,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990916735268198,
      "loss": 5.6404,
      "mean_token_accuracy": 0.1422861024737358,
      "num_tokens": 8474566.0,
      "step": 3855
    },
    {
      "entropy": 5.808441305160523,
      "epoch": 0.3830885271933307,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990884902478966,
      "loss": 5.645,
      "mean_token_accuracy": 0.14045381546020508,
      "num_tokens": 8484786.0,
      "step": 3860
    },
    {
      "entropy": 5.6865410804748535,
      "epoch": 0.3835847558554982,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004990853014120619,
      "loss": 5.571,
      "mean_token_accuracy": 0.14521799385547637,
      "num_tokens": 8496416.0,
      "step": 3865
    },
    {
      "entropy": 5.733003807067871,
      "epoch": 0.38408098451766576,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004990821070193944,
      "loss": 5.6045,
      "mean_token_accuracy": 0.14117611795663834,
      "num_tokens": 8507287.0,
      "step": 3870
    },
    {
      "entropy": 5.785660648345948,
      "epoch": 0.3845772131798333,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004990789070699736,
      "loss": 5.6507,
      "mean_token_accuracy": 0.134336905926466,
      "num_tokens": 8517922.0,
      "step": 3875
    },
    {
      "entropy": 5.702974510192871,
      "epoch": 0.3850734418420008,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990757015638787,
      "loss": 5.5823,
      "mean_token_accuracy": 0.1453675426542759,
      "num_tokens": 8529532.0,
      "step": 3880
    },
    {
      "entropy": 5.749368524551391,
      "epoch": 0.3855696705041683,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990724905011894,
      "loss": 5.5415,
      "mean_token_accuracy": 0.14694293439388276,
      "num_tokens": 8540207.0,
      "step": 3885
    },
    {
      "entropy": 5.764535760879516,
      "epoch": 0.3860658991663358,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499069273881985,
      "loss": 5.6017,
      "mean_token_accuracy": 0.1401916190981865,
      "num_tokens": 8551634.0,
      "step": 3890
    },
    {
      "entropy": 5.759525489807129,
      "epoch": 0.38656212782850335,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004990660517063455,
      "loss": 5.65,
      "mean_token_accuracy": 0.13647492751479148,
      "num_tokens": 8563256.0,
      "step": 3895
    },
    {
      "entropy": 5.7427960395812985,
      "epoch": 0.3870583564906709,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004990628239743508,
      "loss": 5.6617,
      "mean_token_accuracy": 0.1428539350628853,
      "num_tokens": 8573833.0,
      "step": 3900
    },
    {
      "entropy": 5.713337182998657,
      "epoch": 0.3875545851528384,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004990595906860807,
      "loss": 5.5842,
      "mean_token_accuracy": 0.1448860026896,
      "num_tokens": 8584283.0,
      "step": 3905
    },
    {
      "entropy": 5.746373891830444,
      "epoch": 0.38805081381500595,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004990563518416157,
      "loss": 5.6359,
      "mean_token_accuracy": 0.140063364058733,
      "num_tokens": 8595344.0,
      "step": 3910
    },
    {
      "entropy": 5.745514059066773,
      "epoch": 0.3885470424771735,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990531074410358,
      "loss": 5.5535,
      "mean_token_accuracy": 0.14560891613364219,
      "num_tokens": 8605625.0,
      "step": 3915
    },
    {
      "entropy": 5.709172630310059,
      "epoch": 0.389043271139341,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990498574844218,
      "loss": 5.5433,
      "mean_token_accuracy": 0.14344219714403153,
      "num_tokens": 8616259.0,
      "step": 3920
    },
    {
      "entropy": 5.773303079605102,
      "epoch": 0.38953949980150854,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004990466019718542,
      "loss": 5.5579,
      "mean_token_accuracy": 0.14597627222537995,
      "num_tokens": 8627179.0,
      "step": 3925
    },
    {
      "entropy": 5.7424726486206055,
      "epoch": 0.39003572846367607,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004990433409034134,
      "loss": 5.6674,
      "mean_token_accuracy": 0.1455268532037735,
      "num_tokens": 8637448.0,
      "step": 3930
    },
    {
      "entropy": 5.678447723388672,
      "epoch": 0.3905319571258436,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990400742791806,
      "loss": 5.5955,
      "mean_token_accuracy": 0.1387963816523552,
      "num_tokens": 8649168.0,
      "step": 3935
    },
    {
      "entropy": 5.773329401016236,
      "epoch": 0.39102818578801113,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004990368020992368,
      "loss": 5.6941,
      "mean_token_accuracy": 0.1367001138627529,
      "num_tokens": 8660653.0,
      "step": 3940
    },
    {
      "entropy": 5.7819396495819095,
      "epoch": 0.39152441445017866,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499033524363663,
      "loss": 5.6639,
      "mean_token_accuracy": 0.14397113472223283,
      "num_tokens": 8672000.0,
      "step": 3945
    },
    {
      "entropy": 5.703604078292846,
      "epoch": 0.3920206431123462,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004990302410725405,
      "loss": 5.5842,
      "mean_token_accuracy": 0.14821155071258546,
      "num_tokens": 8682455.0,
      "step": 3950
    },
    {
      "entropy": 5.719656085968017,
      "epoch": 0.3925168717745137,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004990269522259509,
      "loss": 5.5607,
      "mean_token_accuracy": 0.14346430450677872,
      "num_tokens": 8692975.0,
      "step": 3955
    },
    {
      "entropy": 5.700722599029541,
      "epoch": 0.3930131004366812,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990236578239755,
      "loss": 5.5429,
      "mean_token_accuracy": 0.14142076373100282,
      "num_tokens": 8703569.0,
      "step": 3960
    },
    {
      "entropy": 5.761720275878906,
      "epoch": 0.3935093290988487,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990203578666962,
      "loss": 5.6559,
      "mean_token_accuracy": 0.13742631003260614,
      "num_tokens": 8713734.0,
      "step": 3965
    },
    {
      "entropy": 5.743472719192505,
      "epoch": 0.39400555776101626,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004990170523541948,
      "loss": 5.5357,
      "mean_token_accuracy": 0.14768768101930618,
      "num_tokens": 8723852.0,
      "step": 3970
    },
    {
      "entropy": 5.710507488250732,
      "epoch": 0.3945017864231838,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004990137412865531,
      "loss": 5.6887,
      "mean_token_accuracy": 0.14348620921373367,
      "num_tokens": 8734300.0,
      "step": 3975
    },
    {
      "entropy": 5.805154275894165,
      "epoch": 0.3949980150853513,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990104246638533,
      "loss": 5.7054,
      "mean_token_accuracy": 0.13508821055293083,
      "num_tokens": 8745022.0,
      "step": 3980
    },
    {
      "entropy": 5.774425458908081,
      "epoch": 0.39549424374751885,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990071024861778,
      "loss": 5.5833,
      "mean_token_accuracy": 0.13405360728502275,
      "num_tokens": 8754870.0,
      "step": 3985
    },
    {
      "entropy": 5.83691554069519,
      "epoch": 0.3959904724096864,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004990037747536087,
      "loss": 5.6401,
      "mean_token_accuracy": 0.14253394007682801,
      "num_tokens": 8765536.0,
      "step": 3990
    },
    {
      "entropy": 5.679643011093139,
      "epoch": 0.3964867010718539,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990004414662288,
      "loss": 5.6109,
      "mean_token_accuracy": 0.14047741740942002,
      "num_tokens": 8776898.0,
      "step": 3995
    },
    {
      "entropy": 5.740188264846802,
      "epoch": 0.39698292973402144,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989971026241205,
      "loss": 5.6714,
      "mean_token_accuracy": 0.13848020657896995,
      "num_tokens": 8788176.0,
      "step": 4000
    },
    {
      "entropy": 5.700374555587769,
      "epoch": 0.39747915839618897,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989937582273668,
      "loss": 5.5856,
      "mean_token_accuracy": 0.14035871252417564,
      "num_tokens": 8798939.0,
      "step": 4005
    },
    {
      "entropy": 5.715577459335327,
      "epoch": 0.3979753870583565,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004989904082760506,
      "loss": 5.6308,
      "mean_token_accuracy": 0.14493223130702973,
      "num_tokens": 8809092.0,
      "step": 4010
    },
    {
      "entropy": 5.731601619720459,
      "epoch": 0.39847161572052403,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000498987052770255,
      "loss": 5.549,
      "mean_token_accuracy": 0.14861732870340347,
      "num_tokens": 8819352.0,
      "step": 4015
    },
    {
      "entropy": 5.744062423706055,
      "epoch": 0.39896784438269156,
      "grad_norm": 0.9375,
      "learning_rate": 0.000498983691710063,
      "loss": 5.6741,
      "mean_token_accuracy": 0.1401342362165451,
      "num_tokens": 8829878.0,
      "step": 4020
    },
    {
      "entropy": 5.737670612335205,
      "epoch": 0.3994640730448591,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498980325095558,
      "loss": 5.5346,
      "mean_token_accuracy": 0.1425821863114834,
      "num_tokens": 8840364.0,
      "step": 4025
    },
    {
      "entropy": 5.703105068206787,
      "epoch": 0.3999603017070266,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004989769529268238,
      "loss": 5.6547,
      "mean_token_accuracy": 0.1329083226621151,
      "num_tokens": 8851340.0,
      "step": 4030
    },
    {
      "entropy": 5.688699102401733,
      "epoch": 0.4004565303691941,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004989735752039436,
      "loss": 5.6232,
      "mean_token_accuracy": 0.13763951659202575,
      "num_tokens": 8862571.0,
      "step": 4035
    },
    {
      "entropy": 5.747498846054077,
      "epoch": 0.4009527590313616,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004989701919270015,
      "loss": 5.6114,
      "mean_token_accuracy": 0.14419316872954369,
      "num_tokens": 8873927.0,
      "step": 4040
    },
    {
      "entropy": 5.835363340377808,
      "epoch": 0.40144898769352916,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989668030960812,
      "loss": 5.8179,
      "mean_token_accuracy": 0.13200713470578193,
      "num_tokens": 8886542.0,
      "step": 4045
    },
    {
      "entropy": 5.795601892471313,
      "epoch": 0.4019452163556967,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989634087112667,
      "loss": 5.5541,
      "mean_token_accuracy": 0.1411711633205414,
      "num_tokens": 8896114.0,
      "step": 4050
    },
    {
      "entropy": 5.712681913375855,
      "epoch": 0.4024414450178642,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004989600087726425,
      "loss": 5.6337,
      "mean_token_accuracy": 0.135611891746521,
      "num_tokens": 8906303.0,
      "step": 4055
    },
    {
      "entropy": 5.780783605575562,
      "epoch": 0.40293767368003175,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989566032802924,
      "loss": 5.666,
      "mean_token_accuracy": 0.13578887283802032,
      "num_tokens": 8916961.0,
      "step": 4060
    },
    {
      "entropy": 5.8231781959533695,
      "epoch": 0.4034339023421993,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004989531922343013,
      "loss": 5.7521,
      "mean_token_accuracy": 0.13754109293222427,
      "num_tokens": 8927700.0,
      "step": 4065
    },
    {
      "entropy": 5.754679441452026,
      "epoch": 0.4039301310043668,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004989497756347534,
      "loss": 5.5834,
      "mean_token_accuracy": 0.1437943920493126,
      "num_tokens": 8939655.0,
      "step": 4070
    },
    {
      "entropy": 5.698803377151489,
      "epoch": 0.40442635966653434,
      "grad_norm": 0.875,
      "learning_rate": 0.0004989463534817339,
      "loss": 5.525,
      "mean_token_accuracy": 0.14996594488620757,
      "num_tokens": 8951238.0,
      "step": 4075
    },
    {
      "entropy": 5.7337535381317135,
      "epoch": 0.40492258832870187,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004989429257753273,
      "loss": 5.5954,
      "mean_token_accuracy": 0.1474266193807125,
      "num_tokens": 8962019.0,
      "step": 4080
    },
    {
      "entropy": 5.738568735122681,
      "epoch": 0.4054188169908694,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004989394925156187,
      "loss": 5.595,
      "mean_token_accuracy": 0.1431464821100235,
      "num_tokens": 8973861.0,
      "step": 4085
    },
    {
      "entropy": 5.689034509658813,
      "epoch": 0.40591504565303693,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004989360537026932,
      "loss": 5.5913,
      "mean_token_accuracy": 0.14298243969678878,
      "num_tokens": 8985981.0,
      "step": 4090
    },
    {
      "entropy": 5.71806526184082,
      "epoch": 0.40641127431520446,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498932609336636,
      "loss": 5.4788,
      "mean_token_accuracy": 0.14892211481928824,
      "num_tokens": 8995318.0,
      "step": 4095
    },
    {
      "entropy": 5.727901840209961,
      "epoch": 0.406907502977372,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004989291594175327,
      "loss": 5.5317,
      "mean_token_accuracy": 0.1424754813313484,
      "num_tokens": 9004679.0,
      "step": 4100
    },
    {
      "entropy": 5.833022928237915,
      "epoch": 0.4074037316395395,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004989257039454689,
      "loss": 5.6549,
      "mean_token_accuracy": 0.14180289059877396,
      "num_tokens": 9016422.0,
      "step": 4105
    },
    {
      "entropy": 5.692260265350342,
      "epoch": 0.40789996030170705,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00049892224292053,
      "loss": 5.5695,
      "mean_token_accuracy": 0.1427057534456253,
      "num_tokens": 9027914.0,
      "step": 4110
    },
    {
      "entropy": 5.702508449554443,
      "epoch": 0.40839618896387453,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004989187763428021,
      "loss": 5.5434,
      "mean_token_accuracy": 0.14706938862800598,
      "num_tokens": 9038701.0,
      "step": 4115
    },
    {
      "entropy": 5.695271348953247,
      "epoch": 0.40889241762604206,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000498915304212371,
      "loss": 5.6123,
      "mean_token_accuracy": 0.13733669966459275,
      "num_tokens": 9050620.0,
      "step": 4120
    },
    {
      "entropy": 5.745882606506347,
      "epoch": 0.4093886462882096,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989118265293228,
      "loss": 5.6348,
      "mean_token_accuracy": 0.143435500562191,
      "num_tokens": 9060900.0,
      "step": 4125
    },
    {
      "entropy": 5.723922824859619,
      "epoch": 0.4098848749503771,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004989083432937439,
      "loss": 5.5599,
      "mean_token_accuracy": 0.1416636064648628,
      "num_tokens": 9072122.0,
      "step": 4130
    },
    {
      "entropy": 5.714425945281983,
      "epoch": 0.41038110361254465,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004989048545057204,
      "loss": 5.5717,
      "mean_token_accuracy": 0.14303382486104965,
      "num_tokens": 9082780.0,
      "step": 4135
    },
    {
      "entropy": 5.707315826416016,
      "epoch": 0.4108773322747122,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004989013601653391,
      "loss": 5.6098,
      "mean_token_accuracy": 0.14234035238623619,
      "num_tokens": 9093518.0,
      "step": 4140
    },
    {
      "entropy": 5.705009508132934,
      "epoch": 0.4113735609368797,
      "grad_norm": 0.875,
      "learning_rate": 0.0004988978602726866,
      "loss": 5.6371,
      "mean_token_accuracy": 0.14491504356265067,
      "num_tokens": 9105035.0,
      "step": 4145
    },
    {
      "entropy": 5.72711033821106,
      "epoch": 0.41186978959904724,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004988943548278494,
      "loss": 5.5522,
      "mean_token_accuracy": 0.14476445242762565,
      "num_tokens": 9117154.0,
      "step": 4150
    },
    {
      "entropy": 5.721778726577758,
      "epoch": 0.4123660182612148,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988908438309149,
      "loss": 5.6244,
      "mean_token_accuracy": 0.14237917512655257,
      "num_tokens": 9128818.0,
      "step": 4155
    },
    {
      "entropy": 5.778127670288086,
      "epoch": 0.4128622469233823,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004988873272819697,
      "loss": 5.6435,
      "mean_token_accuracy": 0.14399504363536836,
      "num_tokens": 9140188.0,
      "step": 4160
    },
    {
      "entropy": 5.867915105819702,
      "epoch": 0.41335847558554983,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988838051811014,
      "loss": 5.7094,
      "mean_token_accuracy": 0.1409727469086647,
      "num_tokens": 9150565.0,
      "step": 4165
    },
    {
      "entropy": 5.710624837875367,
      "epoch": 0.41385470424771736,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498880277528397,
      "loss": 5.5585,
      "mean_token_accuracy": 0.14689609557390212,
      "num_tokens": 9160130.0,
      "step": 4170
    },
    {
      "entropy": 5.71880955696106,
      "epoch": 0.4143509329098849,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004988767443239443,
      "loss": 5.5574,
      "mean_token_accuracy": 0.14629159718751908,
      "num_tokens": 9171740.0,
      "step": 4175
    },
    {
      "entropy": 5.68405818939209,
      "epoch": 0.4148471615720524,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988732055678307,
      "loss": 5.5073,
      "mean_token_accuracy": 0.15795112699270247,
      "num_tokens": 9182459.0,
      "step": 4180
    },
    {
      "entropy": 5.7594013690948485,
      "epoch": 0.41534339023421996,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498869661260144,
      "loss": 5.5727,
      "mean_token_accuracy": 0.14156269729137422,
      "num_tokens": 9192284.0,
      "step": 4185
    },
    {
      "entropy": 5.685678911209107,
      "epoch": 0.41583961889638743,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498866111400972,
      "loss": 5.566,
      "mean_token_accuracy": 0.14145209789276122,
      "num_tokens": 9202700.0,
      "step": 4190
    },
    {
      "entropy": 5.729474592208862,
      "epoch": 0.41633584755855496,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004988625559904029,
      "loss": 5.5868,
      "mean_token_accuracy": 0.13673395067453384,
      "num_tokens": 9213013.0,
      "step": 4195
    },
    {
      "entropy": 5.746359014511109,
      "epoch": 0.4168320762207225,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004988589950285248,
      "loss": 5.6554,
      "mean_token_accuracy": 0.13856094107031822,
      "num_tokens": 9224568.0,
      "step": 4200
    },
    {
      "entropy": 5.674574184417724,
      "epoch": 0.41732830488289,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004988554285154259,
      "loss": 5.5154,
      "mean_token_accuracy": 0.14188847690820694,
      "num_tokens": 9233791.0,
      "step": 4205
    },
    {
      "entropy": 5.743446207046508,
      "epoch": 0.41782453354505755,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004988518564511948,
      "loss": 5.5913,
      "mean_token_accuracy": 0.13892836272716522,
      "num_tokens": 9245048.0,
      "step": 4210
    },
    {
      "entropy": 5.789479351043701,
      "epoch": 0.4183207622072251,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004988482788359202,
      "loss": 5.5935,
      "mean_token_accuracy": 0.14784503281116484,
      "num_tokens": 9256508.0,
      "step": 4215
    },
    {
      "entropy": 5.7150181293487545,
      "epoch": 0.4188169908693926,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004988446956696904,
      "loss": 5.6271,
      "mean_token_accuracy": 0.14091657102108002,
      "num_tokens": 9267871.0,
      "step": 4220
    },
    {
      "entropy": 5.779106283187867,
      "epoch": 0.41931321953156014,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004988411069525947,
      "loss": 5.6809,
      "mean_token_accuracy": 0.1356142893433571,
      "num_tokens": 9279287.0,
      "step": 4225
    },
    {
      "entropy": 5.804977607727051,
      "epoch": 0.4198094481937277,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004988375126847216,
      "loss": 5.6935,
      "mean_token_accuracy": 0.13691136837005616,
      "num_tokens": 9291257.0,
      "step": 4230
    },
    {
      "entropy": 5.812367677688599,
      "epoch": 0.4203056768558952,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004988339128661607,
      "loss": 5.6579,
      "mean_token_accuracy": 0.14066813588142396,
      "num_tokens": 9302590.0,
      "step": 4235
    },
    {
      "entropy": 5.618175888061524,
      "epoch": 0.42080190551806274,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000498830307497001,
      "loss": 5.515,
      "mean_token_accuracy": 0.1433468133211136,
      "num_tokens": 9313174.0,
      "step": 4240
    },
    {
      "entropy": 5.695446538925171,
      "epoch": 0.42129813418023027,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498826696577332,
      "loss": 5.6063,
      "mean_token_accuracy": 0.14157773107290267,
      "num_tokens": 9325034.0,
      "step": 4245
    },
    {
      "entropy": 5.748422145843506,
      "epoch": 0.4217943628423978,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004988230801072432,
      "loss": 5.5873,
      "mean_token_accuracy": 0.14445904046297073,
      "num_tokens": 9336834.0,
      "step": 4250
    },
    {
      "entropy": 5.697594404220581,
      "epoch": 0.4222905915045653,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004988194580868244,
      "loss": 5.5885,
      "mean_token_accuracy": 0.1432207852602005,
      "num_tokens": 9347747.0,
      "step": 4255
    },
    {
      "entropy": 5.68045597076416,
      "epoch": 0.42278682016673286,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004988158305161654,
      "loss": 5.5934,
      "mean_token_accuracy": 0.1437683343887329,
      "num_tokens": 9359186.0,
      "step": 4260
    },
    {
      "entropy": 5.722374105453492,
      "epoch": 0.42328304882890033,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000498812197395356,
      "loss": 5.514,
      "mean_token_accuracy": 0.14643879979848862,
      "num_tokens": 9369485.0,
      "step": 4265
    },
    {
      "entropy": 5.757485628128052,
      "epoch": 0.42377927749106786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988085587244863,
      "loss": 5.6179,
      "mean_token_accuracy": 0.13760613054037094,
      "num_tokens": 9382658.0,
      "step": 4270
    },
    {
      "entropy": 5.686766576766968,
      "epoch": 0.4242755061532354,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004988049145036466,
      "loss": 5.5651,
      "mean_token_accuracy": 0.13786769136786461,
      "num_tokens": 9394151.0,
      "step": 4275
    },
    {
      "entropy": 5.803422689437866,
      "epoch": 0.4247717348154029,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004988012647329273,
      "loss": 5.6641,
      "mean_token_accuracy": 0.13631192743778228,
      "num_tokens": 9405956.0,
      "step": 4280
    },
    {
      "entropy": 5.7161619663238525,
      "epoch": 0.42526796347757045,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000498797609412419,
      "loss": 5.5532,
      "mean_token_accuracy": 0.14562675803899766,
      "num_tokens": 9416889.0,
      "step": 4285
    },
    {
      "entropy": 5.713563346862793,
      "epoch": 0.425764192139738,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004987939485422122,
      "loss": 5.6597,
      "mean_token_accuracy": 0.13954902291297913,
      "num_tokens": 9427602.0,
      "step": 4290
    },
    {
      "entropy": 5.756892251968384,
      "epoch": 0.4262604208019055,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004987902821223976,
      "loss": 5.5697,
      "mean_token_accuracy": 0.14587784707546234,
      "num_tokens": 9438557.0,
      "step": 4295
    },
    {
      "entropy": 5.696601390838623,
      "epoch": 0.42675664946407305,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004987866101530663,
      "loss": 5.6446,
      "mean_token_accuracy": 0.14027416184544564,
      "num_tokens": 9449598.0,
      "step": 4300
    },
    {
      "entropy": 5.736512470245361,
      "epoch": 0.4272528781262406,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004987829326343092,
      "loss": 5.555,
      "mean_token_accuracy": 0.14784253388643265,
      "num_tokens": 9459740.0,
      "step": 4305
    },
    {
      "entropy": 5.7235182762146,
      "epoch": 0.4277491067884081,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987792495662177,
      "loss": 5.5779,
      "mean_token_accuracy": 0.13899921551346778,
      "num_tokens": 9469835.0,
      "step": 4310
    },
    {
      "entropy": 5.7507165431976315,
      "epoch": 0.42824533545057564,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000498775560948883,
      "loss": 5.6549,
      "mean_token_accuracy": 0.14052388891577722,
      "num_tokens": 9480735.0,
      "step": 4315
    },
    {
      "entropy": 5.712874031066894,
      "epoch": 0.42874156411274317,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004987718667823965,
      "loss": 5.538,
      "mean_token_accuracy": 0.14749773293733598,
      "num_tokens": 9491321.0,
      "step": 4320
    },
    {
      "entropy": 5.732460165023804,
      "epoch": 0.4292377927749107,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00049876816706685,
      "loss": 5.6957,
      "mean_token_accuracy": 0.13651810511946677,
      "num_tokens": 9503507.0,
      "step": 4325
    },
    {
      "entropy": 5.802684688568116,
      "epoch": 0.42973402143707823,
      "grad_norm": 1.7734375,
      "learning_rate": 0.000498764461802335,
      "loss": 5.6834,
      "mean_token_accuracy": 0.14579661712050437,
      "num_tokens": 9514404.0,
      "step": 4330
    },
    {
      "entropy": 5.7324159145355225,
      "epoch": 0.43023025009924576,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004987607509889437,
      "loss": 5.5574,
      "mean_token_accuracy": 0.1418476790189743,
      "num_tokens": 9525555.0,
      "step": 4335
    },
    {
      "entropy": 5.744416856765747,
      "epoch": 0.43072647876141323,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004987570346267677,
      "loss": 5.5349,
      "mean_token_accuracy": 0.1448458880186081,
      "num_tokens": 9535084.0,
      "step": 4340
    },
    {
      "entropy": 5.688948249816894,
      "epoch": 0.43122270742358076,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004987533127158996,
      "loss": 5.5599,
      "mean_token_accuracy": 0.14783779084682463,
      "num_tokens": 9546170.0,
      "step": 4345
    },
    {
      "entropy": 5.707106018066407,
      "epoch": 0.4317189360857483,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004987495852564314,
      "loss": 5.5708,
      "mean_token_accuracy": 0.1440203867852688,
      "num_tokens": 9556793.0,
      "step": 4350
    },
    {
      "entropy": 5.728043746948242,
      "epoch": 0.4322151647479158,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004987458522484556,
      "loss": 5.5347,
      "mean_token_accuracy": 0.14598321914672852,
      "num_tokens": 9568114.0,
      "step": 4355
    },
    {
      "entropy": 5.6903989791870115,
      "epoch": 0.43271139341008336,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987421136920648,
      "loss": 5.5325,
      "mean_token_accuracy": 0.14110969603061677,
      "num_tokens": 9579119.0,
      "step": 4360
    },
    {
      "entropy": 5.637863302230835,
      "epoch": 0.4332076220722509,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004987383695873517,
      "loss": 5.5679,
      "mean_token_accuracy": 0.1449605032801628,
      "num_tokens": 9590259.0,
      "step": 4365
    },
    {
      "entropy": 5.7717413902282715,
      "epoch": 0.4337038507344184,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987346199344092,
      "loss": 5.6091,
      "mean_token_accuracy": 0.13644771203398703,
      "num_tokens": 9600431.0,
      "step": 4370
    },
    {
      "entropy": 5.666158962249756,
      "epoch": 0.43420007939658595,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004987308647333303,
      "loss": 5.5863,
      "mean_token_accuracy": 0.14474302157759666,
      "num_tokens": 9610286.0,
      "step": 4375
    },
    {
      "entropy": 5.7676201343536375,
      "epoch": 0.4346963080587535,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004987271039842079,
      "loss": 5.6053,
      "mean_token_accuracy": 0.14020158275961875,
      "num_tokens": 9621148.0,
      "step": 4380
    },
    {
      "entropy": 5.6954185962677,
      "epoch": 0.435192536720921,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004987233376871355,
      "loss": 5.489,
      "mean_token_accuracy": 0.14512064382433892,
      "num_tokens": 9631300.0,
      "step": 4385
    },
    {
      "entropy": 5.684174966812134,
      "epoch": 0.43568876538308854,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004987195658422063,
      "loss": 5.5825,
      "mean_token_accuracy": 0.14152609258890153,
      "num_tokens": 9641986.0,
      "step": 4390
    },
    {
      "entropy": 5.750404691696167,
      "epoch": 0.43618499404525607,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004987157884495141,
      "loss": 5.6017,
      "mean_token_accuracy": 0.14466868340969086,
      "num_tokens": 9652962.0,
      "step": 4395
    },
    {
      "entropy": 5.7012653827667235,
      "epoch": 0.4366812227074236,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004987120055091523,
      "loss": 5.6411,
      "mean_token_accuracy": 0.1366007797420025,
      "num_tokens": 9664084.0,
      "step": 4400
    },
    {
      "entropy": 5.6462547302246096,
      "epoch": 0.43717745136959113,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004987082170212149,
      "loss": 5.5308,
      "mean_token_accuracy": 0.13876494839787484,
      "num_tokens": 9675821.0,
      "step": 4405
    },
    {
      "entropy": 5.772474527359009,
      "epoch": 0.43767368003175866,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004987044229857957,
      "loss": 5.6539,
      "mean_token_accuracy": 0.1392872929573059,
      "num_tokens": 9687964.0,
      "step": 4410
    },
    {
      "entropy": 5.742184209823608,
      "epoch": 0.43816990869392614,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004987006234029889,
      "loss": 5.6274,
      "mean_token_accuracy": 0.14196077436208726,
      "num_tokens": 9699653.0,
      "step": 4415
    },
    {
      "entropy": 5.769703722000122,
      "epoch": 0.43866613735609367,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004986968182728885,
      "loss": 5.6332,
      "mean_token_accuracy": 0.14099507704377173,
      "num_tokens": 9709872.0,
      "step": 4420
    },
    {
      "entropy": 5.763581466674805,
      "epoch": 0.4391623660182612,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986930075955892,
      "loss": 5.5604,
      "mean_token_accuracy": 0.13856668621301652,
      "num_tokens": 9720589.0,
      "step": 4425
    },
    {
      "entropy": 5.648024702072144,
      "epoch": 0.4396585946804287,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004986891913711853,
      "loss": 5.4917,
      "mean_token_accuracy": 0.15007439404726028,
      "num_tokens": 9731862.0,
      "step": 4430
    },
    {
      "entropy": 5.676476955413818,
      "epoch": 0.44015482334259626,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004986853695997715,
      "loss": 5.6447,
      "mean_token_accuracy": 0.1467808410525322,
      "num_tokens": 9743710.0,
      "step": 4435
    },
    {
      "entropy": 5.7441404342651365,
      "epoch": 0.4406510520047638,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986815422814425,
      "loss": 5.5928,
      "mean_token_accuracy": 0.13124571591615677,
      "num_tokens": 9753855.0,
      "step": 4440
    },
    {
      "entropy": 5.832332277297974,
      "epoch": 0.4411472806669313,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004986777094162932,
      "loss": 5.6,
      "mean_token_accuracy": 0.14974353909492494,
      "num_tokens": 9763883.0,
      "step": 4445
    },
    {
      "entropy": 5.690779829025269,
      "epoch": 0.44164350932909885,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986738710044187,
      "loss": 5.5678,
      "mean_token_accuracy": 0.1452265501022339,
      "num_tokens": 9774890.0,
      "step": 4450
    },
    {
      "entropy": 5.675231504440307,
      "epoch": 0.4421397379912664,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004986700270459142,
      "loss": 5.5401,
      "mean_token_accuracy": 0.14680951908230783,
      "num_tokens": 9785249.0,
      "step": 4455
    },
    {
      "entropy": 5.715075492858887,
      "epoch": 0.4426359666534339,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986661775408749,
      "loss": 5.5418,
      "mean_token_accuracy": 0.1484077528119087,
      "num_tokens": 9795723.0,
      "step": 4460
    },
    {
      "entropy": 5.701987457275391,
      "epoch": 0.44313219531560144,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004986623224893965,
      "loss": 5.4956,
      "mean_token_accuracy": 0.14353932589292526,
      "num_tokens": 9807637.0,
      "step": 4465
    },
    {
      "entropy": 5.737668371200561,
      "epoch": 0.44362842397776897,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004986584618915744,
      "loss": 5.617,
      "mean_token_accuracy": 0.14440734907984734,
      "num_tokens": 9820298.0,
      "step": 4470
    },
    {
      "entropy": 5.6621157169342045,
      "epoch": 0.4441246526399365,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986545957475045,
      "loss": 5.4962,
      "mean_token_accuracy": 0.15061585903167723,
      "num_tokens": 9831251.0,
      "step": 4475
    },
    {
      "entropy": 5.658903312683106,
      "epoch": 0.44462088130210403,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986507240572825,
      "loss": 5.5639,
      "mean_token_accuracy": 0.14525441080331802,
      "num_tokens": 9842884.0,
      "step": 4480
    },
    {
      "entropy": 5.655770349502563,
      "epoch": 0.44511710996427156,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004986468468210042,
      "loss": 5.467,
      "mean_token_accuracy": 0.14644730389118193,
      "num_tokens": 9854413.0,
      "step": 4485
    },
    {
      "entropy": 5.705828237533569,
      "epoch": 0.44561333862643904,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004986429640387663,
      "loss": 5.6078,
      "mean_token_accuracy": 0.1486542619764805,
      "num_tokens": 9865771.0,
      "step": 4490
    },
    {
      "entropy": 5.735600233078003,
      "epoch": 0.44610956728860657,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004986390757106648,
      "loss": 5.521,
      "mean_token_accuracy": 0.1486384704709053,
      "num_tokens": 9876717.0,
      "step": 4495
    },
    {
      "entropy": 5.674195289611816,
      "epoch": 0.4466057959507741,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004986351818367959,
      "loss": 5.6052,
      "mean_token_accuracy": 0.14445709586143493,
      "num_tokens": 9887616.0,
      "step": 4500
    },
    {
      "entropy": 5.7188373565673825,
      "epoch": 0.44710202461294163,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986312824172566,
      "loss": 5.5535,
      "mean_token_accuracy": 0.14710604920983314,
      "num_tokens": 9898979.0,
      "step": 4505
    },
    {
      "entropy": 5.7393416404724125,
      "epoch": 0.44759825327510916,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004986273774521433,
      "loss": 5.598,
      "mean_token_accuracy": 0.13431652784347534,
      "num_tokens": 9912158.0,
      "step": 4510
    },
    {
      "entropy": 5.729260540008545,
      "epoch": 0.4480944819372767,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004986234669415529,
      "loss": 5.6112,
      "mean_token_accuracy": 0.13989987969398499,
      "num_tokens": 9923886.0,
      "step": 4515
    },
    {
      "entropy": 5.710967636108398,
      "epoch": 0.4485907105994442,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004986195508855824,
      "loss": 5.568,
      "mean_token_accuracy": 0.14522476345300675,
      "num_tokens": 9934979.0,
      "step": 4520
    },
    {
      "entropy": 5.73377742767334,
      "epoch": 0.44908693926161175,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498615629284329,
      "loss": 5.6753,
      "mean_token_accuracy": 0.13699942901730539,
      "num_tokens": 9944715.0,
      "step": 4525
    },
    {
      "entropy": 5.720676136016846,
      "epoch": 0.4495831679237793,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004986117021378896,
      "loss": 5.5443,
      "mean_token_accuracy": 0.14494104832410812,
      "num_tokens": 9954736.0,
      "step": 4530
    },
    {
      "entropy": 5.709241533279419,
      "epoch": 0.4500793965859468,
      "grad_norm": 0.8125,
      "learning_rate": 0.000498607769446362,
      "loss": 5.6216,
      "mean_token_accuracy": 0.1421697661280632,
      "num_tokens": 9965952.0,
      "step": 4535
    },
    {
      "entropy": 5.6858762264251705,
      "epoch": 0.45057562524811434,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004986038312098435,
      "loss": 5.5741,
      "mean_token_accuracy": 0.14561808109283447,
      "num_tokens": 9976428.0,
      "step": 4540
    },
    {
      "entropy": 5.689672613143921,
      "epoch": 0.4510718539102819,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985998874284319,
      "loss": 5.5091,
      "mean_token_accuracy": 0.15442328453063964,
      "num_tokens": 9986100.0,
      "step": 4545
    },
    {
      "entropy": 5.609327793121338,
      "epoch": 0.4515680825724494,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004985959381022248,
      "loss": 5.5176,
      "mean_token_accuracy": 0.14544908851385116,
      "num_tokens": 9996584.0,
      "step": 4550
    },
    {
      "entropy": 5.65582242012024,
      "epoch": 0.45206431123461693,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004985919832313203,
      "loss": 5.5209,
      "mean_token_accuracy": 0.1447821080684662,
      "num_tokens": 10007792.0,
      "step": 4555
    },
    {
      "entropy": 5.783624315261841,
      "epoch": 0.45256053989678446,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985880228158164,
      "loss": 5.7078,
      "mean_token_accuracy": 0.13523221537470817,
      "num_tokens": 10019615.0,
      "step": 4560
    },
    {
      "entropy": 5.654605865478516,
      "epoch": 0.45305676855895194,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985840568558112,
      "loss": 5.5268,
      "mean_token_accuracy": 0.14684706926345825,
      "num_tokens": 10030223.0,
      "step": 4565
    },
    {
      "entropy": 5.660999774932861,
      "epoch": 0.45355299722111947,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985800853514033,
      "loss": 5.5514,
      "mean_token_accuracy": 0.14329646304249763,
      "num_tokens": 10041522.0,
      "step": 4570
    },
    {
      "entropy": 5.739897918701172,
      "epoch": 0.454049225883287,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004985761083026911,
      "loss": 5.4876,
      "mean_token_accuracy": 0.1519765079021454,
      "num_tokens": 10052847.0,
      "step": 4575
    },
    {
      "entropy": 5.634422731399536,
      "epoch": 0.45454545454545453,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004985721257097731,
      "loss": 5.5151,
      "mean_token_accuracy": 0.14605328142642976,
      "num_tokens": 10062259.0,
      "step": 4580
    },
    {
      "entropy": 5.746643114089966,
      "epoch": 0.45504168320762206,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004985681375727482,
      "loss": 5.6664,
      "mean_token_accuracy": 0.13895911127328872,
      "num_tokens": 10073341.0,
      "step": 4585
    },
    {
      "entropy": 5.727483177185059,
      "epoch": 0.4555379118697896,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004985641438917151,
      "loss": 5.5213,
      "mean_token_accuracy": 0.15104319751262665,
      "num_tokens": 10083992.0,
      "step": 4590
    },
    {
      "entropy": 5.7053868770599365,
      "epoch": 0.4560341405319571,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004985601446667731,
      "loss": 5.5157,
      "mean_token_accuracy": 0.14777568504214286,
      "num_tokens": 10094245.0,
      "step": 4595
    },
    {
      "entropy": 5.688941955566406,
      "epoch": 0.45653036919412465,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004985561398980213,
      "loss": 5.5144,
      "mean_token_accuracy": 0.14640356823801995,
      "num_tokens": 10105976.0,
      "step": 4600
    },
    {
      "entropy": 5.641877317428589,
      "epoch": 0.4570265978562922,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985521295855588,
      "loss": 5.5733,
      "mean_token_accuracy": 0.14555746018886567,
      "num_tokens": 10117373.0,
      "step": 4605
    },
    {
      "entropy": 5.679246425628662,
      "epoch": 0.4575228265184597,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004985481137294853,
      "loss": 5.5183,
      "mean_token_accuracy": 0.13970350474119186,
      "num_tokens": 10128470.0,
      "step": 4610
    },
    {
      "entropy": 5.621714115142822,
      "epoch": 0.45801905518062724,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004985440923299002,
      "loss": 5.5212,
      "mean_token_accuracy": 0.14765125513076782,
      "num_tokens": 10140217.0,
      "step": 4615
    },
    {
      "entropy": 5.667717123031617,
      "epoch": 0.4585152838427948,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004985400653869034,
      "loss": 5.5875,
      "mean_token_accuracy": 0.14465930163860322,
      "num_tokens": 10152308.0,
      "step": 4620
    },
    {
      "entropy": 5.760606670379639,
      "epoch": 0.4590115125049623,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004985360329005945,
      "loss": 5.6278,
      "mean_token_accuracy": 0.14379535466432572,
      "num_tokens": 10164234.0,
      "step": 4625
    },
    {
      "entropy": 5.850100708007813,
      "epoch": 0.45950774116712984,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985319948710738,
      "loss": 5.6696,
      "mean_token_accuracy": 0.14368383213877678,
      "num_tokens": 10175287.0,
      "step": 4630
    },
    {
      "entropy": 5.714134311676025,
      "epoch": 0.46000396982929737,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985279512984413,
      "loss": 5.5401,
      "mean_token_accuracy": 0.1474534623324871,
      "num_tokens": 10184658.0,
      "step": 4635
    },
    {
      "entropy": 5.683081340789795,
      "epoch": 0.46050019849146484,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985239021827972,
      "loss": 5.5407,
      "mean_token_accuracy": 0.14281928688287734,
      "num_tokens": 10194593.0,
      "step": 4640
    },
    {
      "entropy": 5.643809127807617,
      "epoch": 0.46099642715363237,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000498519847524242,
      "loss": 5.4609,
      "mean_token_accuracy": 0.1482139438390732,
      "num_tokens": 10205200.0,
      "step": 4645
    },
    {
      "entropy": 5.627954339981079,
      "epoch": 0.4614926558157999,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004985157873228762,
      "loss": 5.4901,
      "mean_token_accuracy": 0.1481429785490036,
      "num_tokens": 10216848.0,
      "step": 4650
    },
    {
      "entropy": 5.701125526428223,
      "epoch": 0.46198888447796743,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985117215788005,
      "loss": 5.5382,
      "mean_token_accuracy": 0.1468523286283016,
      "num_tokens": 10228188.0,
      "step": 4655
    },
    {
      "entropy": 5.6984175682067875,
      "epoch": 0.46248511314013496,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004985076502921158,
      "loss": 5.6165,
      "mean_token_accuracy": 0.14205382913351058,
      "num_tokens": 10238253.0,
      "step": 4660
    },
    {
      "entropy": 5.732058429718018,
      "epoch": 0.4629813418023025,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498503573462923,
      "loss": 5.5538,
      "mean_token_accuracy": 0.1446946606040001,
      "num_tokens": 10248169.0,
      "step": 4665
    },
    {
      "entropy": 5.6297468662261965,
      "epoch": 0.46347757046447,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004984994910913232,
      "loss": 5.4295,
      "mean_token_accuracy": 0.1478232428431511,
      "num_tokens": 10258962.0,
      "step": 4670
    },
    {
      "entropy": 5.6839498519897464,
      "epoch": 0.46397379912663755,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004984954031774175,
      "loss": 5.5555,
      "mean_token_accuracy": 0.14447656720876695,
      "num_tokens": 10270303.0,
      "step": 4675
    },
    {
      "entropy": 5.606175422668457,
      "epoch": 0.4644700277888051,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004984913097213075,
      "loss": 5.4721,
      "mean_token_accuracy": 0.14679303541779518,
      "num_tokens": 10281782.0,
      "step": 4680
    },
    {
      "entropy": 5.665231561660766,
      "epoch": 0.4649662564509726,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004984872107230945,
      "loss": 5.5035,
      "mean_token_accuracy": 0.14818502962589264,
      "num_tokens": 10292937.0,
      "step": 4685
    },
    {
      "entropy": 5.675636720657349,
      "epoch": 0.46546248511314015,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004984831061828803,
      "loss": 5.5788,
      "mean_token_accuracy": 0.14178428724408149,
      "num_tokens": 10305443.0,
      "step": 4690
    },
    {
      "entropy": 5.661796236038208,
      "epoch": 0.4659587137753077,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004984789961007667,
      "loss": 5.5021,
      "mean_token_accuracy": 0.14874687641859055,
      "num_tokens": 10316489.0,
      "step": 4695
    },
    {
      "entropy": 5.598919534683228,
      "epoch": 0.4664549424374752,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004984748804768554,
      "loss": 5.5602,
      "mean_token_accuracy": 0.13966515585780143,
      "num_tokens": 10326690.0,
      "step": 4700
    },
    {
      "entropy": 5.724484539031982,
      "epoch": 0.46695117109964274,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004984707593112487,
      "loss": 5.548,
      "mean_token_accuracy": 0.14317874610424042,
      "num_tokens": 10337059.0,
      "step": 4705
    },
    {
      "entropy": 5.647740983963013,
      "epoch": 0.46744739976181027,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004984666326040487,
      "loss": 5.4308,
      "mean_token_accuracy": 0.1530276544392109,
      "num_tokens": 10347035.0,
      "step": 4710
    },
    {
      "entropy": 5.6334373474121096,
      "epoch": 0.46794362842397774,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004984625003553578,
      "loss": 5.5579,
      "mean_token_accuracy": 0.14425093233585357,
      "num_tokens": 10359492.0,
      "step": 4715
    },
    {
      "entropy": 5.715490818023682,
      "epoch": 0.4684398570861453,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004984583625652783,
      "loss": 5.5419,
      "mean_token_accuracy": 0.14544321745634078,
      "num_tokens": 10370128.0,
      "step": 4720
    },
    {
      "entropy": 5.689985609054565,
      "epoch": 0.4689360857483128,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000498454219233913,
      "loss": 5.5389,
      "mean_token_accuracy": 0.14008901715278627,
      "num_tokens": 10380276.0,
      "step": 4725
    },
    {
      "entropy": 5.613158082962036,
      "epoch": 0.46943231441048033,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004984500703613645,
      "loss": 5.5232,
      "mean_token_accuracy": 0.14934480264782907,
      "num_tokens": 10392489.0,
      "step": 4730
    },
    {
      "entropy": 5.769829893112183,
      "epoch": 0.46992854307264786,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004984459159477358,
      "loss": 5.6114,
      "mean_token_accuracy": 0.14083144292235375,
      "num_tokens": 10404259.0,
      "step": 4735
    },
    {
      "entropy": 5.626959705352784,
      "epoch": 0.4704247717348154,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004984417559931298,
      "loss": 5.474,
      "mean_token_accuracy": 0.14718540757894516,
      "num_tokens": 10415000.0,
      "step": 4740
    },
    {
      "entropy": 5.70870304107666,
      "epoch": 0.4709210003969829,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984375904976498,
      "loss": 5.5996,
      "mean_token_accuracy": 0.1425722323358059,
      "num_tokens": 10426078.0,
      "step": 4745
    },
    {
      "entropy": 5.705552864074707,
      "epoch": 0.47141722905915046,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004984334194613989,
      "loss": 5.5221,
      "mean_token_accuracy": 0.14522773921489715,
      "num_tokens": 10436838.0,
      "step": 4750
    },
    {
      "entropy": 5.6620279312133786,
      "epoch": 0.471913457721318,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004984292428844809,
      "loss": 5.5104,
      "mean_token_accuracy": 0.1477622404694557,
      "num_tokens": 10448544.0,
      "step": 4755
    },
    {
      "entropy": 5.723327922821045,
      "epoch": 0.4724096863834855,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004984250607669989,
      "loss": 5.5926,
      "mean_token_accuracy": 0.13789047300815582,
      "num_tokens": 10459730.0,
      "step": 4760
    },
    {
      "entropy": 5.760179853439331,
      "epoch": 0.47290591504565305,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004984208731090569,
      "loss": 5.6464,
      "mean_token_accuracy": 0.14240211620926857,
      "num_tokens": 10470257.0,
      "step": 4765
    },
    {
      "entropy": 5.725982236862182,
      "epoch": 0.4734021437078206,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004984166799107587,
      "loss": 5.5736,
      "mean_token_accuracy": 0.1431686021387577,
      "num_tokens": 10480835.0,
      "step": 4770
    },
    {
      "entropy": 5.673721694946289,
      "epoch": 0.4738983723699881,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004984124811722081,
      "loss": 5.6174,
      "mean_token_accuracy": 0.14570664018392562,
      "num_tokens": 10491720.0,
      "step": 4775
    },
    {
      "entropy": 5.677192306518554,
      "epoch": 0.47439460103215564,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004984082768935096,
      "loss": 5.6288,
      "mean_token_accuracy": 0.14744448885321618,
      "num_tokens": 10502313.0,
      "step": 4780
    },
    {
      "entropy": 5.717935991287232,
      "epoch": 0.47489082969432317,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004984040670747672,
      "loss": 5.5059,
      "mean_token_accuracy": 0.14801656156778337,
      "num_tokens": 10513333.0,
      "step": 4785
    },
    {
      "entropy": 5.642219924926758,
      "epoch": 0.47538705835649064,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004983998517160852,
      "loss": 5.4789,
      "mean_token_accuracy": 0.15268410742282867,
      "num_tokens": 10522766.0,
      "step": 4790
    },
    {
      "entropy": 5.641929483413696,
      "epoch": 0.4758832870186582,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004983956308175684,
      "loss": 5.5273,
      "mean_token_accuracy": 0.1454133540391922,
      "num_tokens": 10534362.0,
      "step": 4795
    },
    {
      "entropy": 5.717424297332764,
      "epoch": 0.4763795156808257,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983914043793212,
      "loss": 5.5404,
      "mean_token_accuracy": 0.14541754499077797,
      "num_tokens": 10545072.0,
      "step": 4800
    },
    {
      "entropy": 5.662140846252441,
      "epoch": 0.47687574434299324,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004983871724014487,
      "loss": 5.4779,
      "mean_token_accuracy": 0.14503563791513444,
      "num_tokens": 10555953.0,
      "step": 4805
    },
    {
      "entropy": 5.675953578948975,
      "epoch": 0.47737197300516077,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004983829348840556,
      "loss": 5.5266,
      "mean_token_accuracy": 0.15017574280500412,
      "num_tokens": 10567757.0,
      "step": 4810
    },
    {
      "entropy": 5.5973015308380125,
      "epoch": 0.4778682016673283,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000498378691827247,
      "loss": 5.4211,
      "mean_token_accuracy": 0.14696255847811698,
      "num_tokens": 10579390.0,
      "step": 4815
    },
    {
      "entropy": 5.703798151016235,
      "epoch": 0.4783644303294958,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004983744432311282,
      "loss": 5.6241,
      "mean_token_accuracy": 0.1413818195462227,
      "num_tokens": 10589624.0,
      "step": 4820
    },
    {
      "entropy": 5.770223808288574,
      "epoch": 0.47886065899166336,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004983701890958046,
      "loss": 5.6033,
      "mean_token_accuracy": 0.13691395223140718,
      "num_tokens": 10601927.0,
      "step": 4825
    },
    {
      "entropy": 5.598169279098511,
      "epoch": 0.4793568876538309,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004983659294213815,
      "loss": 5.4757,
      "mean_token_accuracy": 0.15195037573575973,
      "num_tokens": 10614073.0,
      "step": 4830
    },
    {
      "entropy": 5.658059740066529,
      "epoch": 0.4798531163159984,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983616642079649,
      "loss": 5.5083,
      "mean_token_accuracy": 0.14958997219800949,
      "num_tokens": 10625535.0,
      "step": 4835
    },
    {
      "entropy": 5.777277183532715,
      "epoch": 0.48034934497816595,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004983573934556601,
      "loss": 5.4825,
      "mean_token_accuracy": 0.15245484858751296,
      "num_tokens": 10637160.0,
      "step": 4840
    },
    {
      "entropy": 5.7017354488372805,
      "epoch": 0.4808455736403335,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983531171645734,
      "loss": 5.5988,
      "mean_token_accuracy": 0.14110324680805206,
      "num_tokens": 10649061.0,
      "step": 4845
    },
    {
      "entropy": 5.562698173522949,
      "epoch": 0.481341802302501,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004983488353348106,
      "loss": 5.437,
      "mean_token_accuracy": 0.15287891477346421,
      "num_tokens": 10661130.0,
      "step": 4850
    },
    {
      "entropy": 5.650864458084106,
      "epoch": 0.48183803096466854,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004983445479664779,
      "loss": 5.4996,
      "mean_token_accuracy": 0.151998171210289,
      "num_tokens": 10673017.0,
      "step": 4855
    },
    {
      "entropy": 5.735723400115967,
      "epoch": 0.48233425962683607,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004983402550596817,
      "loss": 5.5697,
      "mean_token_accuracy": 0.14529768973588944,
      "num_tokens": 10683268.0,
      "step": 4860
    },
    {
      "entropy": 5.782104110717773,
      "epoch": 0.48283048828900355,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004983359566145285,
      "loss": 5.5835,
      "mean_token_accuracy": 0.14277169853448868,
      "num_tokens": 10694220.0,
      "step": 4865
    },
    {
      "entropy": 5.681487417221069,
      "epoch": 0.4833267169511711,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983316526311248,
      "loss": 5.4965,
      "mean_token_accuracy": 0.14053841307759285,
      "num_tokens": 10704829.0,
      "step": 4870
    },
    {
      "entropy": 5.690288496017456,
      "epoch": 0.4838229456133386,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983273431095773,
      "loss": 5.6949,
      "mean_token_accuracy": 0.1336314469575882,
      "num_tokens": 10717160.0,
      "step": 4875
    },
    {
      "entropy": 5.73415470123291,
      "epoch": 0.48431917427550614,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004983230280499929,
      "loss": 5.498,
      "mean_token_accuracy": 0.14990181624889373,
      "num_tokens": 10728638.0,
      "step": 4880
    },
    {
      "entropy": 5.670762395858764,
      "epoch": 0.48481540293767367,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004983187074524787,
      "loss": 5.5515,
      "mean_token_accuracy": 0.14588793516159057,
      "num_tokens": 10740122.0,
      "step": 4885
    },
    {
      "entropy": 5.650857067108154,
      "epoch": 0.4853116315998412,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004983143813171418,
      "loss": 5.5085,
      "mean_token_accuracy": 0.1518013060092926,
      "num_tokens": 10751103.0,
      "step": 4890
    },
    {
      "entropy": 5.625285959243774,
      "epoch": 0.48580786026200873,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004983100496440895,
      "loss": 5.4702,
      "mean_token_accuracy": 0.15078884214162827,
      "num_tokens": 10761937.0,
      "step": 4895
    },
    {
      "entropy": 5.783539056777954,
      "epoch": 0.48630408892417626,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498305712433429,
      "loss": 5.6993,
      "mean_token_accuracy": 0.13927533775568007,
      "num_tokens": 10773527.0,
      "step": 4900
    },
    {
      "entropy": 5.689421606063843,
      "epoch": 0.4868003175863438,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000498301369685268,
      "loss": 5.5252,
      "mean_token_accuracy": 0.14703643843531608,
      "num_tokens": 10783873.0,
      "step": 4905
    },
    {
      "entropy": 5.613701868057251,
      "epoch": 0.4872965462485113,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004982970213997142,
      "loss": 5.381,
      "mean_token_accuracy": 0.15223184525966643,
      "num_tokens": 10796124.0,
      "step": 4910
    },
    {
      "entropy": 5.588850116729736,
      "epoch": 0.48779277491067885,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004982926675768756,
      "loss": 5.49,
      "mean_token_accuracy": 0.14737663865089418,
      "num_tokens": 10806468.0,
      "step": 4915
    },
    {
      "entropy": 5.650877809524536,
      "epoch": 0.4882890035728464,
      "grad_norm": 0.875,
      "learning_rate": 0.0004982883082168599,
      "loss": 5.5443,
      "mean_token_accuracy": 0.14648173600435258,
      "num_tokens": 10817126.0,
      "step": 4920
    },
    {
      "entropy": 5.713279438018799,
      "epoch": 0.4887852322350139,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004982839433197752,
      "loss": 5.5715,
      "mean_token_accuracy": 0.1365367963910103,
      "num_tokens": 10828985.0,
      "step": 4925
    },
    {
      "entropy": 5.649182891845703,
      "epoch": 0.48928146089718144,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00049827957288573,
      "loss": 5.4752,
      "mean_token_accuracy": 0.14631946086883546,
      "num_tokens": 10840528.0,
      "step": 4930
    },
    {
      "entropy": 5.659715747833252,
      "epoch": 0.48977768955934897,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004982751969148324,
      "loss": 5.4823,
      "mean_token_accuracy": 0.1495818853378296,
      "num_tokens": 10851312.0,
      "step": 4935
    },
    {
      "entropy": 5.6238807201385494,
      "epoch": 0.49027391822151645,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982708154071912,
      "loss": 5.4437,
      "mean_token_accuracy": 0.14900542199611663,
      "num_tokens": 10861882.0,
      "step": 4940
    },
    {
      "entropy": 5.668314647674561,
      "epoch": 0.490770146883684,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004982664283629147,
      "loss": 5.5941,
      "mean_token_accuracy": 0.13935748636722564,
      "num_tokens": 10873187.0,
      "step": 4945
    },
    {
      "entropy": 5.6875807762146,
      "epoch": 0.4912663755458515,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004982620357821118,
      "loss": 5.4848,
      "mean_token_accuracy": 0.15269453823566437,
      "num_tokens": 10884254.0,
      "step": 4950
    },
    {
      "entropy": 5.59837589263916,
      "epoch": 0.49176260420801904,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004982576376648918,
      "loss": 5.4791,
      "mean_token_accuracy": 0.14832527041435242,
      "num_tokens": 10895770.0,
      "step": 4955
    },
    {
      "entropy": 5.6609045505523685,
      "epoch": 0.49225883287018657,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004982532340113633,
      "loss": 5.4942,
      "mean_token_accuracy": 0.1478537067770958,
      "num_tokens": 10906436.0,
      "step": 4960
    },
    {
      "entropy": 5.698761129379273,
      "epoch": 0.4927550615323541,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004982488248216357,
      "loss": 5.4663,
      "mean_token_accuracy": 0.14812514334917068,
      "num_tokens": 10917655.0,
      "step": 4965
    },
    {
      "entropy": 5.648922395706177,
      "epoch": 0.49325129019452163,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004982444100958182,
      "loss": 5.4702,
      "mean_token_accuracy": 0.15194573104381562,
      "num_tokens": 10928359.0,
      "step": 4970
    },
    {
      "entropy": 5.663105535507202,
      "epoch": 0.49374751885668916,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004982399898340205,
      "loss": 5.637,
      "mean_token_accuracy": 0.13940111473202704,
      "num_tokens": 10940494.0,
      "step": 4975
    },
    {
      "entropy": 5.739015007019043,
      "epoch": 0.4942437475188567,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000498235564036352,
      "loss": 5.6336,
      "mean_token_accuracy": 0.1354987032711506,
      "num_tokens": 10951298.0,
      "step": 4980
    },
    {
      "entropy": 5.670893716812134,
      "epoch": 0.4947399761810242,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004982311327029226,
      "loss": 5.5232,
      "mean_token_accuracy": 0.14640430435538293,
      "num_tokens": 10961680.0,
      "step": 4985
    },
    {
      "entropy": 5.775196647644043,
      "epoch": 0.49523620484319175,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004982266958338421,
      "loss": 5.6519,
      "mean_token_accuracy": 0.14135513976216316,
      "num_tokens": 10974968.0,
      "step": 4990
    },
    {
      "entropy": 5.642757892608643,
      "epoch": 0.4957324335053593,
      "grad_norm": 0.875,
      "learning_rate": 0.0004982222534292205,
      "loss": 5.5701,
      "mean_token_accuracy": 0.1384653314948082,
      "num_tokens": 10986383.0,
      "step": 4995
    },
    {
      "entropy": 5.758541297912598,
      "epoch": 0.4962286621675268,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004982178054891681,
      "loss": 5.6179,
      "mean_token_accuracy": 0.14474987238645554,
      "num_tokens": 10998384.0,
      "step": 5000
    },
    {
      "entropy": 5.681232976913452,
      "epoch": 0.49672489082969434,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004982133520137951,
      "loss": 5.389,
      "mean_token_accuracy": 0.15461313128471374,
      "num_tokens": 11009420.0,
      "step": 5005
    },
    {
      "entropy": 5.6752293586730955,
      "epoch": 0.4972211194918619,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004982088930032119,
      "loss": 5.5462,
      "mean_token_accuracy": 0.14680904522538185,
      "num_tokens": 11019601.0,
      "step": 5010
    },
    {
      "entropy": 5.649302768707275,
      "epoch": 0.4977173481540294,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004982044284575293,
      "loss": 5.4825,
      "mean_token_accuracy": 0.15077524483203888,
      "num_tokens": 11031138.0,
      "step": 5015
    },
    {
      "entropy": 5.6932378768920895,
      "epoch": 0.4982135768161969,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004981999583768576,
      "loss": 5.4972,
      "mean_token_accuracy": 0.14869402647018432,
      "num_tokens": 11041643.0,
      "step": 5020
    },
    {
      "entropy": 5.639556217193603,
      "epoch": 0.4987098054783644,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000498195482761308,
      "loss": 5.5166,
      "mean_token_accuracy": 0.13713015839457512,
      "num_tokens": 11052116.0,
      "step": 5025
    },
    {
      "entropy": 5.763868427276611,
      "epoch": 0.49920603414053194,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004981910016109914,
      "loss": 5.6962,
      "mean_token_accuracy": 0.1334000639617443,
      "num_tokens": 11064796.0,
      "step": 5030
    },
    {
      "entropy": 5.795920133590698,
      "epoch": 0.49970226280269947,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004981865149260189,
      "loss": 5.5422,
      "mean_token_accuracy": 0.14919748455286025,
      "num_tokens": 11075029.0,
      "step": 5035
    },
    {
      "entropy": 5.691865539550781,
      "epoch": 0.5001984914648671,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004981820227065016,
      "loss": 5.5688,
      "mean_token_accuracy": 0.1439378082752228,
      "num_tokens": 11086702.0,
      "step": 5040
    },
    {
      "entropy": 5.636436605453492,
      "epoch": 0.5006947201270345,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004981775249525513,
      "loss": 5.5487,
      "mean_token_accuracy": 0.14065493866801262,
      "num_tokens": 11097417.0,
      "step": 5045
    },
    {
      "entropy": 5.619754123687744,
      "epoch": 0.5011909487892021,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004981730216642791,
      "loss": 5.4312,
      "mean_token_accuracy": 0.15412285625934602,
      "num_tokens": 11108458.0,
      "step": 5050
    },
    {
      "entropy": 5.71159348487854,
      "epoch": 0.5016871774513696,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004981685128417969,
      "loss": 5.606,
      "mean_token_accuracy": 0.14178437143564224,
      "num_tokens": 11121466.0,
      "step": 5055
    },
    {
      "entropy": 5.693905544281006,
      "epoch": 0.5021834061135371,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004981639984852165,
      "loss": 5.5735,
      "mean_token_accuracy": 0.14030421599745752,
      "num_tokens": 11132489.0,
      "step": 5060
    },
    {
      "entropy": 5.643926811218262,
      "epoch": 0.5026796347757047,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004981594785946498,
      "loss": 5.5035,
      "mean_token_accuracy": 0.14792548716068268,
      "num_tokens": 11143087.0,
      "step": 5065
    },
    {
      "entropy": 5.711865997314453,
      "epoch": 0.5031758634378721,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004981549531702089,
      "loss": 5.5463,
      "mean_token_accuracy": 0.13861528635025025,
      "num_tokens": 11154229.0,
      "step": 5070
    },
    {
      "entropy": 5.684705591201782,
      "epoch": 0.5036720921000397,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000498150422212006,
      "loss": 5.5091,
      "mean_token_accuracy": 0.1434071958065033,
      "num_tokens": 11165698.0,
      "step": 5075
    },
    {
      "entropy": 5.68370566368103,
      "epoch": 0.5041683207622072,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004981458857201534,
      "loss": 5.5029,
      "mean_token_accuracy": 0.14430436715483666,
      "num_tokens": 11177025.0,
      "step": 5080
    },
    {
      "entropy": 5.708558464050293,
      "epoch": 0.5046645494243748,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004981413436947639,
      "loss": 5.5667,
      "mean_token_accuracy": 0.14835076332092284,
      "num_tokens": 11188766.0,
      "step": 5085
    },
    {
      "entropy": 5.664322233200073,
      "epoch": 0.5051607780865423,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004981367961359496,
      "loss": 5.6283,
      "mean_token_accuracy": 0.13773046880960466,
      "num_tokens": 11201626.0,
      "step": 5090
    },
    {
      "entropy": 5.6853649616241455,
      "epoch": 0.5056570067487098,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004981322430438239,
      "loss": 5.533,
      "mean_token_accuracy": 0.14464087933301925,
      "num_tokens": 11213975.0,
      "step": 5095
    },
    {
      "entropy": 5.66510968208313,
      "epoch": 0.5061532354108773,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004981276844184991,
      "loss": 5.4853,
      "mean_token_accuracy": 0.1474761888384819,
      "num_tokens": 11223636.0,
      "step": 5100
    },
    {
      "entropy": 5.611086797714234,
      "epoch": 0.5066494640730449,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004981231202600886,
      "loss": 5.4442,
      "mean_token_accuracy": 0.14874403178691864,
      "num_tokens": 11233549.0,
      "step": 5105
    },
    {
      "entropy": 5.596583366394043,
      "epoch": 0.5071456927352124,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004981185505687056,
      "loss": 5.4526,
      "mean_token_accuracy": 0.1505897492170334,
      "num_tokens": 11242915.0,
      "step": 5110
    },
    {
      "entropy": 5.740023183822632,
      "epoch": 0.50764192139738,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004981139753444632,
      "loss": 5.5914,
      "mean_token_accuracy": 0.1402211457490921,
      "num_tokens": 11254370.0,
      "step": 5115
    },
    {
      "entropy": 5.672887802124023,
      "epoch": 0.5081381500595474,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004981093945874752,
      "loss": 5.5167,
      "mean_token_accuracy": 0.1479219064116478,
      "num_tokens": 11264226.0,
      "step": 5120
    },
    {
      "entropy": 5.613948059082031,
      "epoch": 0.508634378721715,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004981048082978547,
      "loss": 5.4124,
      "mean_token_accuracy": 0.15597307235002517,
      "num_tokens": 11275304.0,
      "step": 5125
    },
    {
      "entropy": 5.612861156463623,
      "epoch": 0.5091306073838825,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004981002164757158,
      "loss": 5.4731,
      "mean_token_accuracy": 0.14841337352991105,
      "num_tokens": 11286179.0,
      "step": 5130
    },
    {
      "entropy": 5.707505702972412,
      "epoch": 0.50962683604605,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004980956191211723,
      "loss": 5.5044,
      "mean_token_accuracy": 0.15050283074378967,
      "num_tokens": 11297244.0,
      "step": 5135
    },
    {
      "entropy": 5.691849660873413,
      "epoch": 0.5101230647082176,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004980910162343382,
      "loss": 5.5116,
      "mean_token_accuracy": 0.15071954280138017,
      "num_tokens": 11307251.0,
      "step": 5140
    },
    {
      "entropy": 5.67484941482544,
      "epoch": 0.510619293370385,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004980864078153275,
      "loss": 5.4943,
      "mean_token_accuracy": 0.14468513876199723,
      "num_tokens": 11318441.0,
      "step": 5145
    },
    {
      "entropy": 5.574022626876831,
      "epoch": 0.5111155220325526,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980817938642548,
      "loss": 5.4672,
      "mean_token_accuracy": 0.1504348635673523,
      "num_tokens": 11328743.0,
      "step": 5150
    },
    {
      "entropy": 5.621407842636108,
      "epoch": 0.5116117506947201,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000498077174381234,
      "loss": 5.4571,
      "mean_token_accuracy": 0.1529311530292034,
      "num_tokens": 11340745.0,
      "step": 5155
    },
    {
      "entropy": 5.703346395492554,
      "epoch": 0.5121079793568877,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004980725493663802,
      "loss": 5.5203,
      "mean_token_accuracy": 0.1512206144630909,
      "num_tokens": 11352060.0,
      "step": 5160
    },
    {
      "entropy": 5.640616035461425,
      "epoch": 0.5126042080190552,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004980679188198078,
      "loss": 5.5279,
      "mean_token_accuracy": 0.14537407457828522,
      "num_tokens": 11363097.0,
      "step": 5165
    },
    {
      "entropy": 5.687601852416992,
      "epoch": 0.5131004366812227,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980632827416317,
      "loss": 5.5476,
      "mean_token_accuracy": 0.16172647774219512,
      "num_tokens": 11373410.0,
      "step": 5170
    },
    {
      "entropy": 5.705497169494629,
      "epoch": 0.5135966653433902,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004980586411319669,
      "loss": 5.6674,
      "mean_token_accuracy": 0.14129337221384047,
      "num_tokens": 11385615.0,
      "step": 5175
    },
    {
      "entropy": 5.601473951339722,
      "epoch": 0.5140928940055578,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004980539939909283,
      "loss": 5.4292,
      "mean_token_accuracy": 0.14421194940805435,
      "num_tokens": 11396492.0,
      "step": 5180
    },
    {
      "entropy": 5.699062871932983,
      "epoch": 0.5145891226677253,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004980493413186314,
      "loss": 5.5396,
      "mean_token_accuracy": 0.1392931617796421,
      "num_tokens": 11406377.0,
      "step": 5185
    },
    {
      "entropy": 5.685783576965332,
      "epoch": 0.5150853513298929,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004980446831151914,
      "loss": 5.5101,
      "mean_token_accuracy": 0.14487480521202087,
      "num_tokens": 11416862.0,
      "step": 5190
    },
    {
      "entropy": 5.63937292098999,
      "epoch": 0.5155815799920603,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980400193807239,
      "loss": 5.4585,
      "mean_token_accuracy": 0.15452056229114533,
      "num_tokens": 11428246.0,
      "step": 5195
    },
    {
      "entropy": 5.656797409057617,
      "epoch": 0.5160778086542279,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980353501153446,
      "loss": 5.4988,
      "mean_token_accuracy": 0.1489669419825077,
      "num_tokens": 11438147.0,
      "step": 5200
    },
    {
      "entropy": 5.71633825302124,
      "epoch": 0.5165740373163954,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004980306753191691,
      "loss": 5.6226,
      "mean_token_accuracy": 0.14095016047358513,
      "num_tokens": 11449154.0,
      "step": 5205
    },
    {
      "entropy": 5.693401622772217,
      "epoch": 0.5170702659785629,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004980259949923135,
      "loss": 5.5201,
      "mean_token_accuracy": 0.14495897442102432,
      "num_tokens": 11459262.0,
      "step": 5210
    },
    {
      "entropy": 5.697604036331176,
      "epoch": 0.5175664946407305,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004980213091348938,
      "loss": 5.5126,
      "mean_token_accuracy": 0.14773540049791337,
      "num_tokens": 11471991.0,
      "step": 5215
    },
    {
      "entropy": 5.627869701385498,
      "epoch": 0.5180627233028979,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980166177470262,
      "loss": 5.4505,
      "mean_token_accuracy": 0.14557459205389023,
      "num_tokens": 11483176.0,
      "step": 5220
    },
    {
      "entropy": 5.701073884963989,
      "epoch": 0.5185589519650655,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004980119208288271,
      "loss": 5.5483,
      "mean_token_accuracy": 0.14131275564432144,
      "num_tokens": 11494015.0,
      "step": 5225
    },
    {
      "entropy": 5.613106107711792,
      "epoch": 0.519055180627233,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004980072183804128,
      "loss": 5.4654,
      "mean_token_accuracy": 0.1503124602138996,
      "num_tokens": 11506365.0,
      "step": 5230
    },
    {
      "entropy": 5.626119470596313,
      "epoch": 0.5195514092894006,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004980025104019,
      "loss": 5.5456,
      "mean_token_accuracy": 0.15200978443026542,
      "num_tokens": 11517199.0,
      "step": 5235
    },
    {
      "entropy": 5.624682331085205,
      "epoch": 0.520047637951568,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004979977968934056,
      "loss": 5.3987,
      "mean_token_accuracy": 0.15699853301048278,
      "num_tokens": 11527591.0,
      "step": 5240
    },
    {
      "entropy": 5.696583080291748,
      "epoch": 0.5205438666137356,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004979930778550463,
      "loss": 5.5699,
      "mean_token_accuracy": 0.14902450740337372,
      "num_tokens": 11538657.0,
      "step": 5245
    },
    {
      "entropy": 5.607371425628662,
      "epoch": 0.5210400952759031,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004979883532869392,
      "loss": 5.3807,
      "mean_token_accuracy": 0.16424401253461837,
      "num_tokens": 11549513.0,
      "step": 5250
    },
    {
      "entropy": 5.681191349029541,
      "epoch": 0.5215363239380707,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004979836231892013,
      "loss": 5.5036,
      "mean_token_accuracy": 0.1442022740840912,
      "num_tokens": 11560376.0,
      "step": 5255
    },
    {
      "entropy": 5.589213800430298,
      "epoch": 0.5220325526002382,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979788875619501,
      "loss": 5.4602,
      "mean_token_accuracy": 0.14428113251924515,
      "num_tokens": 11571218.0,
      "step": 5260
    },
    {
      "entropy": 5.65407133102417,
      "epoch": 0.5225287812624058,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000497974146405303,
      "loss": 5.5262,
      "mean_token_accuracy": 0.14889778792858124,
      "num_tokens": 11581981.0,
      "step": 5265
    },
    {
      "entropy": 5.662479019165039,
      "epoch": 0.5230250099245732,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979693997193776,
      "loss": 5.4796,
      "mean_token_accuracy": 0.14792142063379288,
      "num_tokens": 11592561.0,
      "step": 5270
    },
    {
      "entropy": 5.681849956512451,
      "epoch": 0.5235212385867408,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004979646475042914,
      "loss": 5.4594,
      "mean_token_accuracy": 0.14910868704319,
      "num_tokens": 11601933.0,
      "step": 5275
    },
    {
      "entropy": 5.601380777359009,
      "epoch": 0.5240174672489083,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004979598897601625,
      "loss": 5.3937,
      "mean_token_accuracy": 0.1597262978553772,
      "num_tokens": 11613938.0,
      "step": 5280
    },
    {
      "entropy": 5.693992376327515,
      "epoch": 0.5245136959110758,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004979551264871088,
      "loss": 5.5608,
      "mean_token_accuracy": 0.14049421474337578,
      "num_tokens": 11624890.0,
      "step": 5285
    },
    {
      "entropy": 5.656209802627563,
      "epoch": 0.5250099245732434,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004979503576852482,
      "loss": 5.5634,
      "mean_token_accuracy": 0.14738093614578246,
      "num_tokens": 11635080.0,
      "step": 5290
    },
    {
      "entropy": 5.559004783630371,
      "epoch": 0.5255061532354108,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979455833546994,
      "loss": 5.4507,
      "mean_token_accuracy": 0.150251816958189,
      "num_tokens": 11645180.0,
      "step": 5295
    },
    {
      "entropy": 5.698576164245606,
      "epoch": 0.5260023818975784,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004979408034955804,
      "loss": 5.5122,
      "mean_token_accuracy": 0.1495388776063919,
      "num_tokens": 11657599.0,
      "step": 5300
    },
    {
      "entropy": 5.709982585906983,
      "epoch": 0.5264986105597459,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004979360181080099,
      "loss": 5.554,
      "mean_token_accuracy": 0.13851185664534568,
      "num_tokens": 11669177.0,
      "step": 5305
    },
    {
      "entropy": 5.611147880554199,
      "epoch": 0.5269948392219135,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004979312271921065,
      "loss": 5.4022,
      "mean_token_accuracy": 0.14745137840509415,
      "num_tokens": 11679756.0,
      "step": 5310
    },
    {
      "entropy": 5.605481243133545,
      "epoch": 0.527491067884081,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000497926430747989,
      "loss": 5.4801,
      "mean_token_accuracy": 0.1458248570561409,
      "num_tokens": 11689805.0,
      "step": 5315
    },
    {
      "entropy": 5.648860311508178,
      "epoch": 0.5279872965462485,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004979216287757764,
      "loss": 5.4855,
      "mean_token_accuracy": 0.15177066773176193,
      "num_tokens": 11701150.0,
      "step": 5320
    },
    {
      "entropy": 5.582261037826538,
      "epoch": 0.528483525208416,
      "grad_norm": 0.90625,
      "learning_rate": 0.000497916821275588,
      "loss": 5.4237,
      "mean_token_accuracy": 0.1469492107629776,
      "num_tokens": 11712256.0,
      "step": 5325
    },
    {
      "entropy": 5.678399562835693,
      "epoch": 0.5289797538705836,
      "grad_norm": 0.875,
      "learning_rate": 0.0004979120082475424,
      "loss": 5.4753,
      "mean_token_accuracy": 0.15113273933529853,
      "num_tokens": 11722547.0,
      "step": 5330
    },
    {
      "entropy": 5.63621883392334,
      "epoch": 0.5294759825327511,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004979071896917595,
      "loss": 5.4934,
      "mean_token_accuracy": 0.14984367564320564,
      "num_tokens": 11732309.0,
      "step": 5335
    },
    {
      "entropy": 5.65453200340271,
      "epoch": 0.5299722111949187,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004979023656083586,
      "loss": 5.5385,
      "mean_token_accuracy": 0.1465301111340523,
      "num_tokens": 11743982.0,
      "step": 5340
    },
    {
      "entropy": 5.62791337966919,
      "epoch": 0.5304684398570861,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004978975359974594,
      "loss": 5.4629,
      "mean_token_accuracy": 0.14960571080446244,
      "num_tokens": 11755506.0,
      "step": 5345
    },
    {
      "entropy": 5.659137630462647,
      "epoch": 0.5309646685192537,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978927008591816,
      "loss": 5.4984,
      "mean_token_accuracy": 0.14654770717024804,
      "num_tokens": 11767024.0,
      "step": 5350
    },
    {
      "entropy": 5.666571617126465,
      "epoch": 0.5314608971814212,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004978878601936451,
      "loss": 5.4781,
      "mean_token_accuracy": 0.14480792880058288,
      "num_tokens": 11777625.0,
      "step": 5355
    },
    {
      "entropy": 5.589890098571777,
      "epoch": 0.5319571258435887,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004978830140009698,
      "loss": 5.464,
      "mean_token_accuracy": 0.15269244760274886,
      "num_tokens": 11788610.0,
      "step": 5360
    },
    {
      "entropy": 5.667603826522827,
      "epoch": 0.5324533545057563,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004978781622812762,
      "loss": 5.5754,
      "mean_token_accuracy": 0.14783891588449477,
      "num_tokens": 11800828.0,
      "step": 5365
    },
    {
      "entropy": 5.670748043060303,
      "epoch": 0.5329495831679237,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004978733050346844,
      "loss": 5.5546,
      "mean_token_accuracy": 0.15211703032255172,
      "num_tokens": 11811729.0,
      "step": 5370
    },
    {
      "entropy": 5.684667110443115,
      "epoch": 0.5334458118300913,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004978684422613149,
      "loss": 5.5888,
      "mean_token_accuracy": 0.14408276826143265,
      "num_tokens": 11822081.0,
      "step": 5375
    },
    {
      "entropy": 5.630998611450195,
      "epoch": 0.5339420404922588,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004978635739612882,
      "loss": 5.4549,
      "mean_token_accuracy": 0.15072351321578026,
      "num_tokens": 11832984.0,
      "step": 5380
    },
    {
      "entropy": 5.620605039596557,
      "epoch": 0.5344382691544264,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004978587001347252,
      "loss": 5.4317,
      "mean_token_accuracy": 0.14728395640850067,
      "num_tokens": 11842918.0,
      "step": 5385
    },
    {
      "entropy": 5.566420698165894,
      "epoch": 0.5349344978165939,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004978538207817465,
      "loss": 5.4613,
      "mean_token_accuracy": 0.1546679764986038,
      "num_tokens": 11854364.0,
      "step": 5390
    },
    {
      "entropy": 5.636065721511841,
      "epoch": 0.5354307264787614,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978489359024734,
      "loss": 5.551,
      "mean_token_accuracy": 0.1455624908208847,
      "num_tokens": 11866400.0,
      "step": 5395
    },
    {
      "entropy": 5.71275749206543,
      "epoch": 0.5359269551409289,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978440454970268,
      "loss": 5.5318,
      "mean_token_accuracy": 0.14682282507419586,
      "num_tokens": 11877094.0,
      "step": 5400
    },
    {
      "entropy": 5.656115961074829,
      "epoch": 0.5364231838030965,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978391495655281,
      "loss": 5.4814,
      "mean_token_accuracy": 0.15012079179286958,
      "num_tokens": 11887289.0,
      "step": 5405
    },
    {
      "entropy": 5.633537435531617,
      "epoch": 0.536919412465264,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978342481080987,
      "loss": 5.4713,
      "mean_token_accuracy": 0.14486898928880693,
      "num_tokens": 11897700.0,
      "step": 5410
    },
    {
      "entropy": 5.528839206695556,
      "epoch": 0.5374156411274316,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004978293411248601,
      "loss": 5.3919,
      "mean_token_accuracy": 0.15333716124296187,
      "num_tokens": 11908623.0,
      "step": 5415
    },
    {
      "entropy": 5.619808053970337,
      "epoch": 0.537911869789599,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004978244286159339,
      "loss": 5.4142,
      "mean_token_accuracy": 0.15683804154396058,
      "num_tokens": 11917883.0,
      "step": 5420
    },
    {
      "entropy": 5.594229030609131,
      "epoch": 0.5384080984517666,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004978195105814421,
      "loss": 5.4572,
      "mean_token_accuracy": 0.15008229464292527,
      "num_tokens": 11929145.0,
      "step": 5425
    },
    {
      "entropy": 5.644812965393067,
      "epoch": 0.5389043271139341,
      "grad_norm": 0.875,
      "learning_rate": 0.0004978145870215065,
      "loss": 5.4511,
      "mean_token_accuracy": 0.14874640107154846,
      "num_tokens": 11940073.0,
      "step": 5430
    },
    {
      "entropy": 5.658492660522461,
      "epoch": 0.5394005557761016,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004978096579362493,
      "loss": 5.5777,
      "mean_token_accuracy": 0.14769978895783425,
      "num_tokens": 11953140.0,
      "step": 5435
    },
    {
      "entropy": 5.670812654495239,
      "epoch": 0.5398967844382692,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004978047233257927,
      "loss": 5.4993,
      "mean_token_accuracy": 0.15172669440507888,
      "num_tokens": 11963853.0,
      "step": 5440
    },
    {
      "entropy": 5.62924633026123,
      "epoch": 0.5403930131004366,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004977997831902591,
      "loss": 5.4477,
      "mean_token_accuracy": 0.15352184772491456,
      "num_tokens": 11975469.0,
      "step": 5445
    },
    {
      "entropy": 5.611998891830444,
      "epoch": 0.5408892417626042,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004977948375297709,
      "loss": 5.4473,
      "mean_token_accuracy": 0.15664440840482713,
      "num_tokens": 11985227.0,
      "step": 5450
    },
    {
      "entropy": 5.523879051208496,
      "epoch": 0.5413854704247717,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004977898863444509,
      "loss": 5.4524,
      "mean_token_accuracy": 0.14888169020414352,
      "num_tokens": 11995459.0,
      "step": 5455
    },
    {
      "entropy": 5.634063768386841,
      "epoch": 0.5418816990869393,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004977849296344217,
      "loss": 5.4949,
      "mean_token_accuracy": 0.1459165871143341,
      "num_tokens": 12007705.0,
      "step": 5460
    },
    {
      "entropy": 5.597906303405762,
      "epoch": 0.5423779277491068,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004977799673998063,
      "loss": 5.3994,
      "mean_token_accuracy": 0.1513359174132347,
      "num_tokens": 12019729.0,
      "step": 5465
    },
    {
      "entropy": 5.621879291534424,
      "epoch": 0.5428741564112743,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004977749996407278,
      "loss": 5.4778,
      "mean_token_accuracy": 0.14614856690168382,
      "num_tokens": 12031009.0,
      "step": 5470
    },
    {
      "entropy": 5.717142868041992,
      "epoch": 0.5433703850734418,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004977700263573094,
      "loss": 5.4436,
      "mean_token_accuracy": 0.15253398865461348,
      "num_tokens": 12041396.0,
      "step": 5475
    },
    {
      "entropy": 5.6496118068695065,
      "epoch": 0.5438666137356094,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004977650475496743,
      "loss": 5.5907,
      "mean_token_accuracy": 0.14329303279519082,
      "num_tokens": 12053112.0,
      "step": 5480
    },
    {
      "entropy": 5.657518005371093,
      "epoch": 0.5443628423977769,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004977600632179459,
      "loss": 5.4678,
      "mean_token_accuracy": 0.14574399664998056,
      "num_tokens": 12064823.0,
      "step": 5485
    },
    {
      "entropy": 5.6894450187683105,
      "epoch": 0.5448590710599445,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004977550733622482,
      "loss": 5.5912,
      "mean_token_accuracy": 0.14535179287195205,
      "num_tokens": 12076656.0,
      "step": 5490
    },
    {
      "entropy": 5.602209186553955,
      "epoch": 0.5453552997221119,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004977500779827045,
      "loss": 5.3786,
      "mean_token_accuracy": 0.1520494595170021,
      "num_tokens": 12086324.0,
      "step": 5495
    },
    {
      "entropy": 5.543206691741943,
      "epoch": 0.5458515283842795,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004977450770794389,
      "loss": 5.4208,
      "mean_token_accuracy": 0.14765701517462732,
      "num_tokens": 12098076.0,
      "step": 5500
    },
    {
      "entropy": 5.603697919845581,
      "epoch": 0.546347757046447,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004977400706525755,
      "loss": 5.4906,
      "mean_token_accuracy": 0.1446567229926586,
      "num_tokens": 12111282.0,
      "step": 5505
    },
    {
      "entropy": 5.623009395599365,
      "epoch": 0.5468439857086145,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004977350587022381,
      "loss": 5.4193,
      "mean_token_accuracy": 0.15579552501440047,
      "num_tokens": 12121932.0,
      "step": 5510
    },
    {
      "entropy": 5.595358753204346,
      "epoch": 0.5473402143707821,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004977300412285513,
      "loss": 5.4218,
      "mean_token_accuracy": 0.15426837503910065,
      "num_tokens": 12133040.0,
      "step": 5515
    },
    {
      "entropy": 5.622896909713745,
      "epoch": 0.5478364430329495,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004977250182316394,
      "loss": 5.485,
      "mean_token_accuracy": 0.15093854367733,
      "num_tokens": 12142577.0,
      "step": 5520
    },
    {
      "entropy": 5.6181727886199955,
      "epoch": 0.5483326716951171,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497719989711627,
      "loss": 5.4525,
      "mean_token_accuracy": 0.15297684520483018,
      "num_tokens": 12153941.0,
      "step": 5525
    },
    {
      "entropy": 5.608196783065796,
      "epoch": 0.5488289003572846,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004977149556686387,
      "loss": 5.5174,
      "mean_token_accuracy": 0.14787946194410323,
      "num_tokens": 12165347.0,
      "step": 5530
    },
    {
      "entropy": 5.545637702941894,
      "epoch": 0.5493251290194522,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004977099161027996,
      "loss": 5.3864,
      "mean_token_accuracy": 0.15185628309845925,
      "num_tokens": 12175705.0,
      "step": 5535
    },
    {
      "entropy": 5.652411890029907,
      "epoch": 0.5498213576816197,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004977048710142342,
      "loss": 5.4364,
      "mean_token_accuracy": 0.1537328541278839,
      "num_tokens": 12186781.0,
      "step": 5540
    },
    {
      "entropy": 5.666394376754761,
      "epoch": 0.5503175863437872,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004976998204030681,
      "loss": 5.4772,
      "mean_token_accuracy": 0.14764580726623536,
      "num_tokens": 12197145.0,
      "step": 5545
    },
    {
      "entropy": 5.549248790740966,
      "epoch": 0.5508138150059547,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004976947642694262,
      "loss": 5.4741,
      "mean_token_accuracy": 0.15212984159588813,
      "num_tokens": 12207637.0,
      "step": 5550
    },
    {
      "entropy": 5.556308317184448,
      "epoch": 0.5513100436681223,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004976897026134341,
      "loss": 5.3988,
      "mean_token_accuracy": 0.15345845073461534,
      "num_tokens": 12218156.0,
      "step": 5555
    },
    {
      "entropy": 5.653855180740356,
      "epoch": 0.5518062723302898,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004976846354352172,
      "loss": 5.4957,
      "mean_token_accuracy": 0.14872472286224364,
      "num_tokens": 12229987.0,
      "step": 5560
    },
    {
      "entropy": 5.643626880645752,
      "epoch": 0.5523025009924574,
      "grad_norm": 0.875,
      "learning_rate": 0.0004976795627349012,
      "loss": 5.4659,
      "mean_token_accuracy": 0.14570514112710953,
      "num_tokens": 12240508.0,
      "step": 5565
    },
    {
      "entropy": 5.559814167022705,
      "epoch": 0.5527987296546248,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004976744845126118,
      "loss": 5.4069,
      "mean_token_accuracy": 0.15187161713838576,
      "num_tokens": 12250317.0,
      "step": 5570
    },
    {
      "entropy": 5.577573823928833,
      "epoch": 0.5532949583167924,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004976694007684751,
      "loss": 5.4525,
      "mean_token_accuracy": 0.15375591665506363,
      "num_tokens": 12262104.0,
      "step": 5575
    },
    {
      "entropy": 5.699751091003418,
      "epoch": 0.5537911869789599,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000497664311502617,
      "loss": 5.4941,
      "mean_token_accuracy": 0.15318164080381394,
      "num_tokens": 12273259.0,
      "step": 5580
    },
    {
      "entropy": 5.682411003112793,
      "epoch": 0.5542874156411274,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004976592167151638,
      "loss": 5.5331,
      "mean_token_accuracy": 0.14391404539346694,
      "num_tokens": 12285504.0,
      "step": 5585
    },
    {
      "entropy": 5.644524049758911,
      "epoch": 0.554783644303295,
      "grad_norm": 1.0,
      "learning_rate": 0.0004976541164062418,
      "loss": 5.4584,
      "mean_token_accuracy": 0.14769638627767562,
      "num_tokens": 12296512.0,
      "step": 5590
    },
    {
      "entropy": 5.512802314758301,
      "epoch": 0.5552798729654624,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004976490105759774,
      "loss": 5.3308,
      "mean_token_accuracy": 0.15871942639350892,
      "num_tokens": 12307948.0,
      "step": 5595
    },
    {
      "entropy": 5.612875652313233,
      "epoch": 0.55577610162763,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004976438992244976,
      "loss": 5.5017,
      "mean_token_accuracy": 0.14463188201189042,
      "num_tokens": 12320020.0,
      "step": 5600
    },
    {
      "entropy": 5.6500156879425045,
      "epoch": 0.5562723302897975,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004976387823519286,
      "loss": 5.3592,
      "mean_token_accuracy": 0.15429917871952056,
      "num_tokens": 12331091.0,
      "step": 5605
    },
    {
      "entropy": 5.6125414848327635,
      "epoch": 0.5567685589519651,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004976336599583975,
      "loss": 5.4673,
      "mean_token_accuracy": 0.14630457013845444,
      "num_tokens": 12342621.0,
      "step": 5610
    },
    {
      "entropy": 5.741844367980957,
      "epoch": 0.5572647876141326,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004976285320440315,
      "loss": 5.7035,
      "mean_token_accuracy": 0.13585652485489846,
      "num_tokens": 12353330.0,
      "step": 5615
    },
    {
      "entropy": 5.693860006332398,
      "epoch": 0.5577610162763001,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004976233986089577,
      "loss": 5.4623,
      "mean_token_accuracy": 0.14677181020379065,
      "num_tokens": 12364418.0,
      "step": 5620
    },
    {
      "entropy": 5.577788209915161,
      "epoch": 0.5582572449384676,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004976182596533031,
      "loss": 5.4472,
      "mean_token_accuracy": 0.15234579965472222,
      "num_tokens": 12377022.0,
      "step": 5625
    },
    {
      "entropy": 5.659402227401733,
      "epoch": 0.5587534736006352,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004976131151771955,
      "loss": 5.5228,
      "mean_token_accuracy": 0.14480556398630143,
      "num_tokens": 12387799.0,
      "step": 5630
    },
    {
      "entropy": 5.5746238231658936,
      "epoch": 0.5592497022628027,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004976079651807623,
      "loss": 5.4466,
      "mean_token_accuracy": 0.14236423075199128,
      "num_tokens": 12397387.0,
      "step": 5635
    },
    {
      "entropy": 5.55435004234314,
      "epoch": 0.5597459309249703,
      "grad_norm": 0.875,
      "learning_rate": 0.0004976028096641313,
      "loss": 5.3789,
      "mean_token_accuracy": 0.15377479195594787,
      "num_tokens": 12407994.0,
      "step": 5640
    },
    {
      "entropy": 5.710486555099488,
      "epoch": 0.5602421595871377,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004975976486274302,
      "loss": 5.5692,
      "mean_token_accuracy": 0.1448943331837654,
      "num_tokens": 12419124.0,
      "step": 5645
    },
    {
      "entropy": 5.650815391540528,
      "epoch": 0.5607383882493053,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004975924820707872,
      "loss": 5.4444,
      "mean_token_accuracy": 0.1529550939798355,
      "num_tokens": 12428699.0,
      "step": 5650
    },
    {
      "entropy": 5.646983575820923,
      "epoch": 0.5612346169114728,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004975873099943302,
      "loss": 5.5266,
      "mean_token_accuracy": 0.14555705338716507,
      "num_tokens": 12439025.0,
      "step": 5655
    },
    {
      "entropy": 5.656699848175049,
      "epoch": 0.5617308455736403,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004975821323981875,
      "loss": 5.4778,
      "mean_token_accuracy": 0.14742110371589662,
      "num_tokens": 12449439.0,
      "step": 5660
    },
    {
      "entropy": 5.654876232147217,
      "epoch": 0.5622270742358079,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004975769492824877,
      "loss": 5.421,
      "mean_token_accuracy": 0.15080877095460893,
      "num_tokens": 12460066.0,
      "step": 5665
    },
    {
      "entropy": 5.6223005771636965,
      "epoch": 0.5627233028979753,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004975717606473589,
      "loss": 5.4397,
      "mean_token_accuracy": 0.15309426784515381,
      "num_tokens": 12471093.0,
      "step": 5670
    },
    {
      "entropy": 5.651199150085449,
      "epoch": 0.5632195315601429,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975665664929302,
      "loss": 5.4579,
      "mean_token_accuracy": 0.1513369768857956,
      "num_tokens": 12481181.0,
      "step": 5675
    },
    {
      "entropy": 5.674373435974121,
      "epoch": 0.5637157602223104,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975613668193303,
      "loss": 5.5064,
      "mean_token_accuracy": 0.14478076249361038,
      "num_tokens": 12493068.0,
      "step": 5680
    },
    {
      "entropy": 5.5631945610046385,
      "epoch": 0.564211988884478,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000497556161626688,
      "loss": 5.4497,
      "mean_token_accuracy": 0.1487413913011551,
      "num_tokens": 12504025.0,
      "step": 5685
    },
    {
      "entropy": 5.607772922515869,
      "epoch": 0.5647082175466455,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004975509509151325,
      "loss": 5.4486,
      "mean_token_accuracy": 0.1523358181118965,
      "num_tokens": 12514567.0,
      "step": 5690
    },
    {
      "entropy": 5.702475070953369,
      "epoch": 0.565204446208813,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975457346847929,
      "loss": 5.4804,
      "mean_token_accuracy": 0.14762943536043166,
      "num_tokens": 12525088.0,
      "step": 5695
    },
    {
      "entropy": 5.618128538131714,
      "epoch": 0.5657006748709805,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004975405129357986,
      "loss": 5.4489,
      "mean_token_accuracy": 0.1503733605146408,
      "num_tokens": 12536002.0,
      "step": 5700
    },
    {
      "entropy": 5.632601690292359,
      "epoch": 0.5661969035331481,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004975352856682792,
      "loss": 5.4669,
      "mean_token_accuracy": 0.15055823028087617,
      "num_tokens": 12547468.0,
      "step": 5705
    },
    {
      "entropy": 5.667331027984619,
      "epoch": 0.5666931321953156,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004975300528823642,
      "loss": 5.4344,
      "mean_token_accuracy": 0.14942139089107515,
      "num_tokens": 12557670.0,
      "step": 5710
    },
    {
      "entropy": 5.5996208667755125,
      "epoch": 0.5671893608574832,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004975248145781834,
      "loss": 5.479,
      "mean_token_accuracy": 0.14795504957437516,
      "num_tokens": 12568425.0,
      "step": 5715
    },
    {
      "entropy": 5.55127854347229,
      "epoch": 0.5676855895196506,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004975195707558667,
      "loss": 5.4463,
      "mean_token_accuracy": 0.14945666939020158,
      "num_tokens": 12579138.0,
      "step": 5720
    },
    {
      "entropy": 5.597843265533447,
      "epoch": 0.5681818181818182,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004975143214155443,
      "loss": 5.4643,
      "mean_token_accuracy": 0.14927031174302102,
      "num_tokens": 12589743.0,
      "step": 5725
    },
    {
      "entropy": 5.586509704589844,
      "epoch": 0.5686780468439857,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975090665573459,
      "loss": 5.447,
      "mean_token_accuracy": 0.15376916229724885,
      "num_tokens": 12600648.0,
      "step": 5730
    },
    {
      "entropy": 5.580662059783935,
      "epoch": 0.5691742755061532,
      "grad_norm": 0.875,
      "learning_rate": 0.0004975038061814023,
      "loss": 5.452,
      "mean_token_accuracy": 0.1524510756134987,
      "num_tokens": 12611369.0,
      "step": 5735
    },
    {
      "entropy": 5.678642797470093,
      "epoch": 0.5696705041683208,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004974985402878437,
      "loss": 5.4363,
      "mean_token_accuracy": 0.1489766277372837,
      "num_tokens": 12621992.0,
      "step": 5740
    },
    {
      "entropy": 5.610474920272827,
      "epoch": 0.5701667328304882,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004974932688768007,
      "loss": 5.5164,
      "mean_token_accuracy": 0.14607659056782724,
      "num_tokens": 12632463.0,
      "step": 5745
    },
    {
      "entropy": 5.548481416702271,
      "epoch": 0.5706629614926558,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004974879919484042,
      "loss": 5.3712,
      "mean_token_accuracy": 0.15110822319984435,
      "num_tokens": 12642282.0,
      "step": 5750
    },
    {
      "entropy": 5.5748039245605465,
      "epoch": 0.5711591901548233,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004974827095027848,
      "loss": 5.4002,
      "mean_token_accuracy": 0.1541622370481491,
      "num_tokens": 12653495.0,
      "step": 5755
    },
    {
      "entropy": 5.586807537078857,
      "epoch": 0.5716554188169909,
      "grad_norm": 0.875,
      "learning_rate": 0.0004974774215400736,
      "loss": 5.3364,
      "mean_token_accuracy": 0.15577918887138367,
      "num_tokens": 12664322.0,
      "step": 5760
    },
    {
      "entropy": 5.507334613800049,
      "epoch": 0.5721516474791584,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004974721280604019,
      "loss": 5.4354,
      "mean_token_accuracy": 0.15019209086894988,
      "num_tokens": 12673981.0,
      "step": 5765
    },
    {
      "entropy": 5.526956558227539,
      "epoch": 0.572647876141326,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004974668290639007,
      "loss": 5.4469,
      "mean_token_accuracy": 0.1492769457399845,
      "num_tokens": 12684778.0,
      "step": 5770
    },
    {
      "entropy": 5.641401767730713,
      "epoch": 0.5731441048034934,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004974615245507014,
      "loss": 5.4059,
      "mean_token_accuracy": 0.1538984641432762,
      "num_tokens": 12695287.0,
      "step": 5775
    },
    {
      "entropy": 5.5436258792877195,
      "epoch": 0.573640333465661,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004974562145209359,
      "loss": 5.4537,
      "mean_token_accuracy": 0.14565087854862213,
      "num_tokens": 12705773.0,
      "step": 5780
    },
    {
      "entropy": 5.538467264175415,
      "epoch": 0.5741365621278285,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004974508989747354,
      "loss": 5.4019,
      "mean_token_accuracy": 0.1574517771601677,
      "num_tokens": 12716783.0,
      "step": 5785
    },
    {
      "entropy": 5.607895183563232,
      "epoch": 0.5746327907899961,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004974455779122321,
      "loss": 5.4516,
      "mean_token_accuracy": 0.14383325651288031,
      "num_tokens": 12728146.0,
      "step": 5790
    },
    {
      "entropy": 5.656476640701294,
      "epoch": 0.5751290194521635,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004974402513335577,
      "loss": 5.4746,
      "mean_token_accuracy": 0.14959257021546363,
      "num_tokens": 12739683.0,
      "step": 5795
    },
    {
      "entropy": 5.6649665355682375,
      "epoch": 0.5756252481143311,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004974349192388444,
      "loss": 5.4589,
      "mean_token_accuracy": 0.14662942439317703,
      "num_tokens": 12750257.0,
      "step": 5800
    },
    {
      "entropy": 5.688394498825073,
      "epoch": 0.5761214767764986,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004974295816282245,
      "loss": 5.5324,
      "mean_token_accuracy": 0.1480068877339363,
      "num_tokens": 12761055.0,
      "step": 5805
    },
    {
      "entropy": 5.610796308517456,
      "epoch": 0.5766177054386661,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974242385018303,
      "loss": 5.4721,
      "mean_token_accuracy": 0.15184674710035323,
      "num_tokens": 12771014.0,
      "step": 5810
    },
    {
      "entropy": 5.567866468429566,
      "epoch": 0.5771139341008337,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000497418889859794,
      "loss": 5.3818,
      "mean_token_accuracy": 0.15442348420619964,
      "num_tokens": 12782623.0,
      "step": 5815
    },
    {
      "entropy": 5.669680786132813,
      "epoch": 0.5776101627630011,
      "grad_norm": 1.0,
      "learning_rate": 0.0004974135357022488,
      "loss": 5.5093,
      "mean_token_accuracy": 0.14822030067443848,
      "num_tokens": 12793533.0,
      "step": 5820
    },
    {
      "entropy": 5.5356302738189695,
      "epoch": 0.5781063914251687,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004974081760293269,
      "loss": 5.3729,
      "mean_token_accuracy": 0.15088124424219132,
      "num_tokens": 12804389.0,
      "step": 5825
    },
    {
      "entropy": 5.58860149383545,
      "epoch": 0.5786026200873362,
      "grad_norm": 0.875,
      "learning_rate": 0.0004974028108411617,
      "loss": 5.4681,
      "mean_token_accuracy": 0.15093735009431838,
      "num_tokens": 12815672.0,
      "step": 5830
    },
    {
      "entropy": 5.639421892166138,
      "epoch": 0.5790988487495038,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973974401378859,
      "loss": 5.4668,
      "mean_token_accuracy": 0.14916255176067353,
      "num_tokens": 12826113.0,
      "step": 5835
    },
    {
      "entropy": 5.580629539489746,
      "epoch": 0.5795950774116713,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004973920639196329,
      "loss": 5.4045,
      "mean_token_accuracy": 0.15356406718492507,
      "num_tokens": 12835941.0,
      "step": 5840
    },
    {
      "entropy": 5.596003293991089,
      "epoch": 0.5800913060738389,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004973866821865358,
      "loss": 5.4899,
      "mean_token_accuracy": 0.14188268631696702,
      "num_tokens": 12846186.0,
      "step": 5845
    },
    {
      "entropy": 5.65446286201477,
      "epoch": 0.5805875347360063,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004973812949387283,
      "loss": 5.4459,
      "mean_token_accuracy": 0.15039265230298043,
      "num_tokens": 12856380.0,
      "step": 5850
    },
    {
      "entropy": 5.568442964553833,
      "epoch": 0.5810837633981739,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973759021763439,
      "loss": 5.4795,
      "mean_token_accuracy": 0.15042850375175476,
      "num_tokens": 12865999.0,
      "step": 5855
    },
    {
      "entropy": 5.599948740005493,
      "epoch": 0.5815799920603414,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973705038995162,
      "loss": 5.3781,
      "mean_token_accuracy": 0.14677538573741913,
      "num_tokens": 12876388.0,
      "step": 5860
    },
    {
      "entropy": 5.578648567199707,
      "epoch": 0.582076220722509,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004973651001083792,
      "loss": 5.3704,
      "mean_token_accuracy": 0.1494818851351738,
      "num_tokens": 12885862.0,
      "step": 5865
    },
    {
      "entropy": 5.555972909927368,
      "epoch": 0.5825724493846764,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004973596908030669,
      "loss": 5.4721,
      "mean_token_accuracy": 0.14565781652927398,
      "num_tokens": 12897489.0,
      "step": 5870
    },
    {
      "entropy": 5.656794214248658,
      "epoch": 0.583068678046844,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004973542759837135,
      "loss": 5.4925,
      "mean_token_accuracy": 0.14666834399104117,
      "num_tokens": 12907395.0,
      "step": 5875
    },
    {
      "entropy": 5.604556083679199,
      "epoch": 0.5835649067090115,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000497348855650453,
      "loss": 5.4358,
      "mean_token_accuracy": 0.15660476684570312,
      "num_tokens": 12919439.0,
      "step": 5880
    },
    {
      "entropy": 5.640682888031006,
      "epoch": 0.584061135371179,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004973434298034201,
      "loss": 5.5249,
      "mean_token_accuracy": 0.14780937358736992,
      "num_tokens": 12931892.0,
      "step": 5885
    },
    {
      "entropy": 5.666601848602295,
      "epoch": 0.5845573640333466,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004973379984427492,
      "loss": 5.5692,
      "mean_token_accuracy": 0.14671001508831977,
      "num_tokens": 12943189.0,
      "step": 5890
    },
    {
      "entropy": 5.634539604187012,
      "epoch": 0.585053592695514,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000497332561568575,
      "loss": 5.5137,
      "mean_token_accuracy": 0.1456242695450783,
      "num_tokens": 12954662.0,
      "step": 5895
    },
    {
      "entropy": 5.624458932876587,
      "epoch": 0.5855498213576816,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004973271191810325,
      "loss": 5.4486,
      "mean_token_accuracy": 0.14609898179769515,
      "num_tokens": 12964924.0,
      "step": 5900
    },
    {
      "entropy": 5.618928527832031,
      "epoch": 0.5860460500198491,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004973216712802564,
      "loss": 5.5248,
      "mean_token_accuracy": 0.14595869034528733,
      "num_tokens": 12976483.0,
      "step": 5905
    },
    {
      "entropy": 5.583964586257935,
      "epoch": 0.5865422786820167,
      "grad_norm": 0.875,
      "learning_rate": 0.000497316217866382,
      "loss": 5.4224,
      "mean_token_accuracy": 0.15400040447711943,
      "num_tokens": 12988054.0,
      "step": 5910
    },
    {
      "entropy": 5.636979436874389,
      "epoch": 0.5870385073441842,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004973107589395444,
      "loss": 5.4794,
      "mean_token_accuracy": 0.15200079306960107,
      "num_tokens": 13000596.0,
      "step": 5915
    },
    {
      "entropy": 5.478622388839722,
      "epoch": 0.5875347360063518,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004973052944998791,
      "loss": 5.3041,
      "mean_token_accuracy": 0.15711194574832915,
      "num_tokens": 13012360.0,
      "step": 5920
    },
    {
      "entropy": 5.5978460788726805,
      "epoch": 0.5880309646685192,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004972998245475214,
      "loss": 5.4745,
      "mean_token_accuracy": 0.14887128621339799,
      "num_tokens": 13023806.0,
      "step": 5925
    },
    {
      "entropy": 5.599380207061768,
      "epoch": 0.5885271933306868,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000497294349082607,
      "loss": 5.3887,
      "mean_token_accuracy": 0.1511357419192791,
      "num_tokens": 13034049.0,
      "step": 5930
    },
    {
      "entropy": 5.611827325820923,
      "epoch": 0.5890234219928543,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004972888681052719,
      "loss": 5.5051,
      "mean_token_accuracy": 0.14897997081279754,
      "num_tokens": 13045006.0,
      "step": 5935
    },
    {
      "entropy": 5.687719774246216,
      "epoch": 0.5895196506550219,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004972833816156518,
      "loss": 5.4084,
      "mean_token_accuracy": 0.14930811673402786,
      "num_tokens": 13055337.0,
      "step": 5940
    },
    {
      "entropy": 5.543897342681885,
      "epoch": 0.5900158793171894,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004972778896138828,
      "loss": 5.4027,
      "mean_token_accuracy": 0.15560031235218047,
      "num_tokens": 13065267.0,
      "step": 5945
    },
    {
      "entropy": 5.578498983383179,
      "epoch": 0.5905121079793569,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972723921001011,
      "loss": 5.467,
      "mean_token_accuracy": 0.15096500515937805,
      "num_tokens": 13075173.0,
      "step": 5950
    },
    {
      "entropy": 5.658969116210938,
      "epoch": 0.5910083366415244,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004972668890744431,
      "loss": 5.4562,
      "mean_token_accuracy": 0.14907519966363908,
      "num_tokens": 13086848.0,
      "step": 5955
    },
    {
      "entropy": 5.581852436065674,
      "epoch": 0.5915045653036919,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004972613805370452,
      "loss": 5.4341,
      "mean_token_accuracy": 0.15141052827239038,
      "num_tokens": 13096865.0,
      "step": 5960
    },
    {
      "entropy": 5.562904024124146,
      "epoch": 0.5920007939658595,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004972558664880439,
      "loss": 5.415,
      "mean_token_accuracy": 0.15210264623165132,
      "num_tokens": 13107907.0,
      "step": 5965
    },
    {
      "entropy": 5.540717506408692,
      "epoch": 0.592497022628027,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000497250346927576,
      "loss": 5.3277,
      "mean_token_accuracy": 0.15800193101167678,
      "num_tokens": 13118171.0,
      "step": 5970
    },
    {
      "entropy": 5.570704555511474,
      "epoch": 0.5929932512901945,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004972448218557785,
      "loss": 5.3144,
      "mean_token_accuracy": 0.16167542934417725,
      "num_tokens": 13127042.0,
      "step": 5975
    },
    {
      "entropy": 5.638939285278321,
      "epoch": 0.593489479952362,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004972392912727882,
      "loss": 5.5744,
      "mean_token_accuracy": 0.15045227780938147,
      "num_tokens": 13137026.0,
      "step": 5980
    },
    {
      "entropy": 5.616703224182129,
      "epoch": 0.5939857086145296,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004972337551787425,
      "loss": 5.4344,
      "mean_token_accuracy": 0.15230552107095718,
      "num_tokens": 13146947.0,
      "step": 5985
    },
    {
      "entropy": 5.668242979049682,
      "epoch": 0.5944819372766971,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004972282135737785,
      "loss": 5.4786,
      "mean_token_accuracy": 0.14680141657590867,
      "num_tokens": 13158714.0,
      "step": 5990
    },
    {
      "entropy": 5.639590787887573,
      "epoch": 0.5949781659388647,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004972226664580337,
      "loss": 5.4931,
      "mean_token_accuracy": 0.1462373726069927,
      "num_tokens": 13170156.0,
      "step": 5995
    },
    {
      "entropy": 5.515760707855224,
      "epoch": 0.5954743946010321,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004972171138316456,
      "loss": 5.3974,
      "mean_token_accuracy": 0.1493689924478531,
      "num_tokens": 13179919.0,
      "step": 6000
    },
    {
      "epoch": 0.5954743946010321,
      "eval_entropy": 5.392710745773623,
      "eval_loss": 5.513803482055664,
      "eval_mean_token_accuracy": 0.15261998519223527,
      "eval_num_tokens": 13179919.0,
      "eval_runtime": 20.7006,
      "eval_samples_per_second": 1626.475,
      "eval_steps_per_second": 203.327,
      "step": 6000
    },
    {
      "entropy": 5.596977090835571,
      "epoch": 0.5959706232631997,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004972115556947518,
      "loss": 5.4779,
      "mean_token_accuracy": 0.1541017025709152,
      "num_tokens": 13190777.0,
      "step": 6005
    },
    {
      "entropy": 5.6757081031799315,
      "epoch": 0.5964668519253672,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004972059920474904,
      "loss": 5.4613,
      "mean_token_accuracy": 0.14994437322020532,
      "num_tokens": 13202179.0,
      "step": 6010
    },
    {
      "entropy": 5.580601167678833,
      "epoch": 0.5969630805875348,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004972004228899991,
      "loss": 5.4181,
      "mean_token_accuracy": 0.1509955532848835,
      "num_tokens": 13212764.0,
      "step": 6015
    },
    {
      "entropy": 5.560248231887817,
      "epoch": 0.5974593092497023,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004971948482224162,
      "loss": 5.4524,
      "mean_token_accuracy": 0.1504641056060791,
      "num_tokens": 13223343.0,
      "step": 6020
    },
    {
      "entropy": 5.629840755462647,
      "epoch": 0.5979555379118698,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004971892680448798,
      "loss": 5.4253,
      "mean_token_accuracy": 0.15569519847631455,
      "num_tokens": 13233839.0,
      "step": 6025
    },
    {
      "entropy": 5.661754989624024,
      "epoch": 0.5984517665740373,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004971836823575283,
      "loss": 5.5301,
      "mean_token_accuracy": 0.14222781658172606,
      "num_tokens": 13243854.0,
      "step": 6030
    },
    {
      "entropy": 5.647662878036499,
      "epoch": 0.5989479952362048,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004971780911605003,
      "loss": 5.4551,
      "mean_token_accuracy": 0.15329526215791703,
      "num_tokens": 13256073.0,
      "step": 6035
    },
    {
      "entropy": 5.599632644653321,
      "epoch": 0.5994442238983724,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004971724944539343,
      "loss": 5.497,
      "mean_token_accuracy": 0.14713055565953254,
      "num_tokens": 13267568.0,
      "step": 6040
    },
    {
      "entropy": 5.649599838256836,
      "epoch": 0.5999404525605398,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004971668922379692,
      "loss": 5.4618,
      "mean_token_accuracy": 0.1436742812395096,
      "num_tokens": 13278829.0,
      "step": 6045
    },
    {
      "entropy": 5.61177659034729,
      "epoch": 0.6004366812227074,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497161284512744,
      "loss": 5.4336,
      "mean_token_accuracy": 0.15138799101114273,
      "num_tokens": 13289156.0,
      "step": 6050
    },
    {
      "entropy": 5.493603658676148,
      "epoch": 0.6009329098848749,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004971556712783976,
      "loss": 5.3112,
      "mean_token_accuracy": 0.1571349173784256,
      "num_tokens": 13299817.0,
      "step": 6055
    },
    {
      "entropy": 5.506628179550171,
      "epoch": 0.6014291385470425,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004971500525350692,
      "loss": 5.3852,
      "mean_token_accuracy": 0.15314638912677764,
      "num_tokens": 13310290.0,
      "step": 6060
    },
    {
      "entropy": 5.533604288101197,
      "epoch": 0.60192536720921,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004971444282828983,
      "loss": 5.4803,
      "mean_token_accuracy": 0.1499824970960617,
      "num_tokens": 13321747.0,
      "step": 6065
    },
    {
      "entropy": 5.6678197383880615,
      "epoch": 0.6024215958713776,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004971387985220242,
      "loss": 5.4208,
      "mean_token_accuracy": 0.1545249879360199,
      "num_tokens": 13333249.0,
      "step": 6070
    },
    {
      "entropy": 5.5247804641723635,
      "epoch": 0.602917824533545,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004971331632525866,
      "loss": 5.4065,
      "mean_token_accuracy": 0.15247410610318185,
      "num_tokens": 13345336.0,
      "step": 6075
    },
    {
      "entropy": 5.669757747650147,
      "epoch": 0.6034140531957126,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000497127522474725,
      "loss": 5.5524,
      "mean_token_accuracy": 0.14289941713213922,
      "num_tokens": 13355797.0,
      "step": 6080
    },
    {
      "entropy": 5.6765429973602295,
      "epoch": 0.6039102818578801,
      "grad_norm": 0.875,
      "learning_rate": 0.0004971218761885797,
      "loss": 5.4941,
      "mean_token_accuracy": 0.14894456416368484,
      "num_tokens": 13367058.0,
      "step": 6085
    },
    {
      "entropy": 5.602103137969971,
      "epoch": 0.6044065105200477,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004971162243942904,
      "loss": 5.4538,
      "mean_token_accuracy": 0.14952910095453262,
      "num_tokens": 13378579.0,
      "step": 6090
    },
    {
      "entropy": 5.651467227935791,
      "epoch": 0.6049027391822152,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004971105670919975,
      "loss": 5.4143,
      "mean_token_accuracy": 0.15067921876907348,
      "num_tokens": 13389240.0,
      "step": 6095
    },
    {
      "entropy": 5.645436096191406,
      "epoch": 0.6053989678443827,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000497104904281841,
      "loss": 5.5716,
      "mean_token_accuracy": 0.14636854827404022,
      "num_tokens": 13401895.0,
      "step": 6100
    },
    {
      "entropy": 5.68249282836914,
      "epoch": 0.6058951965065502,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004970992359639616,
      "loss": 5.5284,
      "mean_token_accuracy": 0.1433717429637909,
      "num_tokens": 13413334.0,
      "step": 6105
    },
    {
      "entropy": 5.65785961151123,
      "epoch": 0.6063914251687177,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004970935621384996,
      "loss": 5.4466,
      "mean_token_accuracy": 0.15460411906242372,
      "num_tokens": 13423371.0,
      "step": 6110
    },
    {
      "entropy": 5.650516653060913,
      "epoch": 0.6068876538308853,
      "grad_norm": 1.890625,
      "learning_rate": 0.0004970878828055959,
      "loss": 5.4274,
      "mean_token_accuracy": 0.15732789784669876,
      "num_tokens": 13434782.0,
      "step": 6115
    },
    {
      "entropy": 5.564152765274048,
      "epoch": 0.6073838824930528,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004970821979653912,
      "loss": 5.414,
      "mean_token_accuracy": 0.14711159765720366,
      "num_tokens": 13444493.0,
      "step": 6120
    },
    {
      "entropy": 5.58135290145874,
      "epoch": 0.6078801111552203,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004970765076180265,
      "loss": 5.4037,
      "mean_token_accuracy": 0.1521766349673271,
      "num_tokens": 13455518.0,
      "step": 6125
    },
    {
      "entropy": 5.663119697570801,
      "epoch": 0.6083763398173878,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004970708117636431,
      "loss": 5.4881,
      "mean_token_accuracy": 0.1530603289604187,
      "num_tokens": 13467631.0,
      "step": 6130
    },
    {
      "entropy": 5.553421258926392,
      "epoch": 0.6088725684795554,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000497065110402382,
      "loss": 5.4247,
      "mean_token_accuracy": 0.14682705998420714,
      "num_tokens": 13478217.0,
      "step": 6135
    },
    {
      "entropy": 5.6026238918304445,
      "epoch": 0.6093687971417229,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970594035343846,
      "loss": 5.405,
      "mean_token_accuracy": 0.15311249345541,
      "num_tokens": 13489169.0,
      "step": 6140
    },
    {
      "entropy": 5.589141130447388,
      "epoch": 0.6098650258038905,
      "grad_norm": 1.0,
      "learning_rate": 0.0004970536911597926,
      "loss": 5.4172,
      "mean_token_accuracy": 0.14844012260437012,
      "num_tokens": 13499273.0,
      "step": 6145
    },
    {
      "entropy": 5.6402270793914795,
      "epoch": 0.6103612544660579,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004970479732787476,
      "loss": 5.461,
      "mean_token_accuracy": 0.15369903594255446,
      "num_tokens": 13509664.0,
      "step": 6150
    },
    {
      "entropy": 5.549899244308472,
      "epoch": 0.6108574831282255,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004970422498913911,
      "loss": 5.3488,
      "mean_token_accuracy": 0.14968640953302384,
      "num_tokens": 13520288.0,
      "step": 6155
    },
    {
      "entropy": 5.564370441436767,
      "epoch": 0.611353711790393,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004970365209978655,
      "loss": 5.4345,
      "mean_token_accuracy": 0.15311274081468582,
      "num_tokens": 13531167.0,
      "step": 6160
    },
    {
      "entropy": 5.590855407714844,
      "epoch": 0.6118499404525606,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004970307865983124,
      "loss": 5.4787,
      "mean_token_accuracy": 0.15183627977967262,
      "num_tokens": 13544109.0,
      "step": 6165
    },
    {
      "entropy": 5.6111372947692875,
      "epoch": 0.6123461691147281,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004970250466928744,
      "loss": 5.4087,
      "mean_token_accuracy": 0.15262484773993493,
      "num_tokens": 13554935.0,
      "step": 6170
    },
    {
      "entropy": 5.625780916213989,
      "epoch": 0.6128423977768956,
      "grad_norm": 0.875,
      "learning_rate": 0.0004970193012816936,
      "loss": 5.4213,
      "mean_token_accuracy": 0.15170982629060745,
      "num_tokens": 13565959.0,
      "step": 6175
    },
    {
      "entropy": 5.550536394119263,
      "epoch": 0.6133386264390631,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004970135503649125,
      "loss": 5.4496,
      "mean_token_accuracy": 0.14747072756290436,
      "num_tokens": 13577883.0,
      "step": 6180
    },
    {
      "entropy": 5.58063268661499,
      "epoch": 0.6138348551012306,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004970077939426738,
      "loss": 5.4037,
      "mean_token_accuracy": 0.15324754267930984,
      "num_tokens": 13588597.0,
      "step": 6185
    },
    {
      "entropy": 5.587204360961914,
      "epoch": 0.6143310837633982,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004970020320151202,
      "loss": 5.3461,
      "mean_token_accuracy": 0.15477967262268066,
      "num_tokens": 13598772.0,
      "step": 6190
    },
    {
      "entropy": 5.640543794631958,
      "epoch": 0.6148273124255657,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004969962645823946,
      "loss": 5.4372,
      "mean_token_accuracy": 0.1521898865699768,
      "num_tokens": 13609696.0,
      "step": 6195
    },
    {
      "entropy": 5.557079124450683,
      "epoch": 0.6153235410877332,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049699049164464,
      "loss": 5.4172,
      "mean_token_accuracy": 0.15375182554125785,
      "num_tokens": 13619389.0,
      "step": 6200
    },
    {
      "entropy": 5.561429595947265,
      "epoch": 0.6158197697499007,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004969847132019995,
      "loss": 5.394,
      "mean_token_accuracy": 0.1538977637887001,
      "num_tokens": 13632133.0,
      "step": 6205
    },
    {
      "entropy": 5.571953344345093,
      "epoch": 0.6163159984120683,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004969789292546164,
      "loss": 5.3465,
      "mean_token_accuracy": 0.1636103093624115,
      "num_tokens": 13642715.0,
      "step": 6210
    },
    {
      "entropy": 5.5559193134307865,
      "epoch": 0.6168122270742358,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969731398026343,
      "loss": 5.4485,
      "mean_token_accuracy": 0.15080342590808868,
      "num_tokens": 13653223.0,
      "step": 6215
    },
    {
      "entropy": 5.517760848999023,
      "epoch": 0.6173084557364034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004969673448461966,
      "loss": 5.4323,
      "mean_token_accuracy": 0.1504549577832222,
      "num_tokens": 13664206.0,
      "step": 6220
    },
    {
      "entropy": 5.590613555908203,
      "epoch": 0.6178046843985708,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000496961544385447,
      "loss": 5.4383,
      "mean_token_accuracy": 0.1557757943868637,
      "num_tokens": 13673974.0,
      "step": 6225
    },
    {
      "entropy": 5.583416986465454,
      "epoch": 0.6183009130607384,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004969557384205295,
      "loss": 5.369,
      "mean_token_accuracy": 0.1549408406019211,
      "num_tokens": 13685189.0,
      "step": 6230
    },
    {
      "entropy": 5.61024580001831,
      "epoch": 0.6187971417229059,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004969499269515878,
      "loss": 5.3969,
      "mean_token_accuracy": 0.15812541991472245,
      "num_tokens": 13696181.0,
      "step": 6235
    },
    {
      "entropy": 5.652813959121704,
      "epoch": 0.6192933703850735,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004969441099787663,
      "loss": 5.5008,
      "mean_token_accuracy": 0.14699973165988922,
      "num_tokens": 13706719.0,
      "step": 6240
    },
    {
      "entropy": 5.647890329360962,
      "epoch": 0.619789599047241,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000496938287502209,
      "loss": 5.452,
      "mean_token_accuracy": 0.15290555059909822,
      "num_tokens": 13716653.0,
      "step": 6245
    },
    {
      "entropy": 5.525570392608643,
      "epoch": 0.6202858277094085,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004969324595220605,
      "loss": 5.3681,
      "mean_token_accuracy": 0.15710528790950776,
      "num_tokens": 13728220.0,
      "step": 6250
    },
    {
      "entropy": 5.50266752243042,
      "epoch": 0.620782056371576,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004969266260384652,
      "loss": 5.332,
      "mean_token_accuracy": 0.15505106896162033,
      "num_tokens": 13738103.0,
      "step": 6255
    },
    {
      "entropy": 5.5926953792572025,
      "epoch": 0.6212782850337435,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004969207870515678,
      "loss": 5.422,
      "mean_token_accuracy": 0.15330670326948165,
      "num_tokens": 13749845.0,
      "step": 6260
    },
    {
      "entropy": 5.616781616210938,
      "epoch": 0.6217745136959111,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000496914942561513,
      "loss": 5.4408,
      "mean_token_accuracy": 0.15307224541902542,
      "num_tokens": 13759983.0,
      "step": 6265
    },
    {
      "entropy": 5.643848037719726,
      "epoch": 0.6222707423580786,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004969090925684459,
      "loss": 5.5296,
      "mean_token_accuracy": 0.1525065764784813,
      "num_tokens": 13770045.0,
      "step": 6270
    },
    {
      "entropy": 5.51793851852417,
      "epoch": 0.6227669710202461,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004969032370725114,
      "loss": 5.3821,
      "mean_token_accuracy": 0.15532010644674302,
      "num_tokens": 13779823.0,
      "step": 6275
    },
    {
      "entropy": 5.583812856674195,
      "epoch": 0.6232631996824136,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004968973760738548,
      "loss": 5.3897,
      "mean_token_accuracy": 0.1501995801925659,
      "num_tokens": 13791165.0,
      "step": 6280
    },
    {
      "entropy": 5.661803197860718,
      "epoch": 0.6237594283445812,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004968915095726216,
      "loss": 5.4912,
      "mean_token_accuracy": 0.14553115367889405,
      "num_tokens": 13802872.0,
      "step": 6285
    },
    {
      "entropy": 5.591525983810425,
      "epoch": 0.6242556570067487,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004968856375689568,
      "loss": 5.4198,
      "mean_token_accuracy": 0.1514755442738533,
      "num_tokens": 13813893.0,
      "step": 6290
    },
    {
      "entropy": 5.59356369972229,
      "epoch": 0.6247518856689163,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004968797600630065,
      "loss": 5.4917,
      "mean_token_accuracy": 0.15136266350746155,
      "num_tokens": 13823613.0,
      "step": 6295
    },
    {
      "entropy": 5.6368231773376465,
      "epoch": 0.6252481143310837,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004968738770549162,
      "loss": 5.4971,
      "mean_token_accuracy": 0.1554287150502205,
      "num_tokens": 13835413.0,
      "step": 6300
    },
    {
      "entropy": 5.605167531967163,
      "epoch": 0.6257443429932513,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004968679885448318,
      "loss": 5.4137,
      "mean_token_accuracy": 0.14931111931800842,
      "num_tokens": 13845806.0,
      "step": 6305
    },
    {
      "entropy": 5.620719861984253,
      "epoch": 0.6262405716554188,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004968620945328995,
      "loss": 5.5048,
      "mean_token_accuracy": 0.15276079922914504,
      "num_tokens": 13856540.0,
      "step": 6310
    },
    {
      "entropy": 5.61713719367981,
      "epoch": 0.6267368003175864,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004968561950192652,
      "loss": 5.3625,
      "mean_token_accuracy": 0.15804028809070586,
      "num_tokens": 13867609.0,
      "step": 6315
    },
    {
      "entropy": 5.62848858833313,
      "epoch": 0.6272330289797539,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004968502900040753,
      "loss": 5.4364,
      "mean_token_accuracy": 0.14521164894104005,
      "num_tokens": 13878299.0,
      "step": 6320
    },
    {
      "entropy": 5.604143857955933,
      "epoch": 0.6277292576419214,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004968443794874764,
      "loss": 5.4484,
      "mean_token_accuracy": 0.15260792821645736,
      "num_tokens": 13889625.0,
      "step": 6325
    },
    {
      "entropy": 5.5830577373504635,
      "epoch": 0.6282254863040889,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004968384634696149,
      "loss": 5.4168,
      "mean_token_accuracy": 0.14463182538747787,
      "num_tokens": 13899680.0,
      "step": 6330
    },
    {
      "entropy": 5.550440120697021,
      "epoch": 0.6287217149662565,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004968325419506375,
      "loss": 5.391,
      "mean_token_accuracy": 0.15744505524635316,
      "num_tokens": 13910883.0,
      "step": 6335
    },
    {
      "entropy": 5.549885511398315,
      "epoch": 0.629217943628424,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004968266149306909,
      "loss": 5.4029,
      "mean_token_accuracy": 0.15215905159711837,
      "num_tokens": 13921442.0,
      "step": 6340
    },
    {
      "entropy": 5.583293914794922,
      "epoch": 0.6297141722905915,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004968206824099224,
      "loss": 5.4343,
      "mean_token_accuracy": 0.15862448066473006,
      "num_tokens": 13930309.0,
      "step": 6345
    },
    {
      "entropy": 5.611448574066162,
      "epoch": 0.630210400952759,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004968147443884789,
      "loss": 5.4261,
      "mean_token_accuracy": 0.15601219236850739,
      "num_tokens": 13941346.0,
      "step": 6350
    },
    {
      "entropy": 5.586319446563721,
      "epoch": 0.6307066296149265,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004968088008665078,
      "loss": 5.396,
      "mean_token_accuracy": 0.1572890430688858,
      "num_tokens": 13951933.0,
      "step": 6355
    },
    {
      "entropy": 5.540216064453125,
      "epoch": 0.6312028582770941,
      "grad_norm": 0.875,
      "learning_rate": 0.0004968028518441563,
      "loss": 5.4616,
      "mean_token_accuracy": 0.1499076098203659,
      "num_tokens": 13963112.0,
      "step": 6360
    },
    {
      "entropy": 5.652147722244263,
      "epoch": 0.6316990869392616,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004967968973215721,
      "loss": 5.4874,
      "mean_token_accuracy": 0.15574448704719543,
      "num_tokens": 13974218.0,
      "step": 6365
    },
    {
      "entropy": 5.657738161087036,
      "epoch": 0.6321953156014292,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004967909372989027,
      "loss": 5.4141,
      "mean_token_accuracy": 0.15234902650117874,
      "num_tokens": 13984841.0,
      "step": 6370
    },
    {
      "entropy": 5.5751903533935545,
      "epoch": 0.6326915442635966,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004967849717762959,
      "loss": 5.4299,
      "mean_token_accuracy": 0.15411732643842696,
      "num_tokens": 13997504.0,
      "step": 6375
    },
    {
      "entropy": 5.528285264968872,
      "epoch": 0.6331877729257642,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004967790007538999,
      "loss": 5.3672,
      "mean_token_accuracy": 0.15952090322971343,
      "num_tokens": 14009244.0,
      "step": 6380
    },
    {
      "entropy": 5.540134716033935,
      "epoch": 0.6336840015879317,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004967730242318624,
      "loss": 5.4008,
      "mean_token_accuracy": 0.15605138689279557,
      "num_tokens": 14021190.0,
      "step": 6385
    },
    {
      "entropy": 5.561956739425659,
      "epoch": 0.6341802302500993,
      "grad_norm": 0.890625,
      "learning_rate": 0.000496767042210332,
      "loss": 5.3639,
      "mean_token_accuracy": 0.156063561886549,
      "num_tokens": 14032191.0,
      "step": 6390
    },
    {
      "entropy": 5.500299978256225,
      "epoch": 0.6346764589122668,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004967610546894566,
      "loss": 5.3296,
      "mean_token_accuracy": 0.16037983745336531,
      "num_tokens": 14042817.0,
      "step": 6395
    },
    {
      "entropy": 5.554322195053101,
      "epoch": 0.6351726875744343,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000496755061669385,
      "loss": 5.3512,
      "mean_token_accuracy": 0.15881706178188323,
      "num_tokens": 14052725.0,
      "step": 6400
    },
    {
      "entropy": 5.610175895690918,
      "epoch": 0.6356689162366018,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004967490631502656,
      "loss": 5.4268,
      "mean_token_accuracy": 0.14564218297600745,
      "num_tokens": 14063621.0,
      "step": 6405
    },
    {
      "entropy": 5.705971002578735,
      "epoch": 0.6361651448987694,
      "grad_norm": 0.875,
      "learning_rate": 0.0004967430591322474,
      "loss": 5.5563,
      "mean_token_accuracy": 0.147991943359375,
      "num_tokens": 14074350.0,
      "step": 6410
    },
    {
      "entropy": 5.67575511932373,
      "epoch": 0.6366613735609369,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004967370496154791,
      "loss": 5.486,
      "mean_token_accuracy": 0.14895145893096923,
      "num_tokens": 14084600.0,
      "step": 6415
    },
    {
      "entropy": 5.625611925125122,
      "epoch": 0.6371576022231044,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004967310346001098,
      "loss": 5.4084,
      "mean_token_accuracy": 0.15440958440303804,
      "num_tokens": 14094739.0,
      "step": 6420
    },
    {
      "entropy": 5.702681303024292,
      "epoch": 0.6376538308852719,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004967250140862885,
      "loss": 5.5624,
      "mean_token_accuracy": 0.14849390238523483,
      "num_tokens": 14106020.0,
      "step": 6425
    },
    {
      "entropy": 5.679549694061279,
      "epoch": 0.6381500595474394,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004967189880741647,
      "loss": 5.5236,
      "mean_token_accuracy": 0.14355285838246346,
      "num_tokens": 14117699.0,
      "step": 6430
    },
    {
      "entropy": 5.57834038734436,
      "epoch": 0.638646288209607,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004967129565638878,
      "loss": 5.4054,
      "mean_token_accuracy": 0.15328127294778823,
      "num_tokens": 14128901.0,
      "step": 6435
    },
    {
      "entropy": 5.588972377777099,
      "epoch": 0.6391425168717745,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004967069195556073,
      "loss": 5.4547,
      "mean_token_accuracy": 0.14977715387940407,
      "num_tokens": 14140832.0,
      "step": 6440
    },
    {
      "entropy": 5.637591886520386,
      "epoch": 0.6396387455339421,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000496700877049473,
      "loss": 5.4916,
      "mean_token_accuracy": 0.15187741667032242,
      "num_tokens": 14151936.0,
      "step": 6445
    },
    {
      "entropy": 5.651413059234619,
      "epoch": 0.6401349741961095,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004966948290456345,
      "loss": 5.4796,
      "mean_token_accuracy": 0.15306032747030257,
      "num_tokens": 14160945.0,
      "step": 6450
    },
    {
      "entropy": 5.560344457626343,
      "epoch": 0.6406312028582771,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004966887755442421,
      "loss": 5.3642,
      "mean_token_accuracy": 0.15178609192371367,
      "num_tokens": 14172052.0,
      "step": 6455
    },
    {
      "entropy": 5.631370782852173,
      "epoch": 0.6411274315204446,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004966827165454456,
      "loss": 5.4448,
      "mean_token_accuracy": 0.1570268116891384,
      "num_tokens": 14181788.0,
      "step": 6460
    },
    {
      "entropy": 5.668099117279053,
      "epoch": 0.6416236601826122,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004966766520493956,
      "loss": 5.5322,
      "mean_token_accuracy": 0.14923309087753295,
      "num_tokens": 14193668.0,
      "step": 6465
    },
    {
      "entropy": 5.659458875656128,
      "epoch": 0.6421198888447797,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004966705820562422,
      "loss": 5.4092,
      "mean_token_accuracy": 0.15575137436389924,
      "num_tokens": 14203665.0,
      "step": 6470
    },
    {
      "entropy": 5.593989849090576,
      "epoch": 0.6426161175069472,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004966645065661361,
      "loss": 5.4438,
      "mean_token_accuracy": 0.1480250433087349,
      "num_tokens": 14214213.0,
      "step": 6475
    },
    {
      "entropy": 5.6232795238494875,
      "epoch": 0.6431123461691147,
      "grad_norm": 1.0,
      "learning_rate": 0.0004966584255792279,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1423339731991291,
      "num_tokens": 14225754.0,
      "step": 6480
    },
    {
      "entropy": 5.638065958023072,
      "epoch": 0.6436085748312823,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004966523390956683,
      "loss": 5.4231,
      "mean_token_accuracy": 0.15263875126838683,
      "num_tokens": 14236508.0,
      "step": 6485
    },
    {
      "entropy": 5.632704401016236,
      "epoch": 0.6441048034934498,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004966462471156083,
      "loss": 5.4648,
      "mean_token_accuracy": 0.1550041601061821,
      "num_tokens": 14247719.0,
      "step": 6490
    },
    {
      "entropy": 5.61082239151001,
      "epoch": 0.6446010321556173,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004966401496391989,
      "loss": 5.5119,
      "mean_token_accuracy": 0.1417395904660225,
      "num_tokens": 14258667.0,
      "step": 6495
    },
    {
      "entropy": 5.609544277191162,
      "epoch": 0.6450972608177848,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004966340466665915,
      "loss": 5.5156,
      "mean_token_accuracy": 0.14988971799612044,
      "num_tokens": 14269666.0,
      "step": 6500
    },
    {
      "entropy": 5.559063959121704,
      "epoch": 0.6455934894799523,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004966279381979373,
      "loss": 5.4094,
      "mean_token_accuracy": 0.15216174125671386,
      "num_tokens": 14280862.0,
      "step": 6505
    },
    {
      "entropy": 5.688417387008667,
      "epoch": 0.6460897181421199,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004966218242333879,
      "loss": 5.5598,
      "mean_token_accuracy": 0.13700098246335984,
      "num_tokens": 14292505.0,
      "step": 6510
    },
    {
      "entropy": 5.637891864776611,
      "epoch": 0.6465859468042874,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004966157047730945,
      "loss": 5.3477,
      "mean_token_accuracy": 0.15833587795495987,
      "num_tokens": 14302547.0,
      "step": 6515
    },
    {
      "entropy": 5.537370729446411,
      "epoch": 0.647082175466455,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004966095798172095,
      "loss": 5.4881,
      "mean_token_accuracy": 0.1493873566389084,
      "num_tokens": 14313585.0,
      "step": 6520
    },
    {
      "entropy": 5.529198169708252,
      "epoch": 0.6475784041286224,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004966034493658843,
      "loss": 5.3499,
      "mean_token_accuracy": 0.15040139853954315,
      "num_tokens": 14323665.0,
      "step": 6525
    },
    {
      "entropy": 5.586475276947022,
      "epoch": 0.64807463279079,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004965973134192711,
      "loss": 5.3843,
      "mean_token_accuracy": 0.1553342416882515,
      "num_tokens": 14334257.0,
      "step": 6530
    },
    {
      "entropy": 5.642743921279907,
      "epoch": 0.6485708614529575,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000496591171977522,
      "loss": 5.4145,
      "mean_token_accuracy": 0.1533318355679512,
      "num_tokens": 14345350.0,
      "step": 6535
    },
    {
      "entropy": 5.611017465591431,
      "epoch": 0.6490670901151251,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004965850250407893,
      "loss": 5.4279,
      "mean_token_accuracy": 0.152386437356472,
      "num_tokens": 14355593.0,
      "step": 6540
    },
    {
      "entropy": 5.564835166931152,
      "epoch": 0.6495633187772926,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004965788726092256,
      "loss": 5.3091,
      "mean_token_accuracy": 0.15809425115585327,
      "num_tokens": 14366405.0,
      "step": 6545
    },
    {
      "entropy": 5.564842176437378,
      "epoch": 0.6500595474394602,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004965727146829832,
      "loss": 5.4131,
      "mean_token_accuracy": 0.1491552025079727,
      "num_tokens": 14377228.0,
      "step": 6550
    },
    {
      "entropy": 5.696504354476929,
      "epoch": 0.6505557761016276,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004965665512622148,
      "loss": 5.5438,
      "mean_token_accuracy": 0.1498422294855118,
      "num_tokens": 14387113.0,
      "step": 6555
    },
    {
      "entropy": 5.623318910598755,
      "epoch": 0.6510520047637952,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004965603823470735,
      "loss": 5.41,
      "mean_token_accuracy": 0.15291822105646133,
      "num_tokens": 14397980.0,
      "step": 6560
    },
    {
      "entropy": 5.563955879211425,
      "epoch": 0.6515482334259627,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004965542079377122,
      "loss": 5.3477,
      "mean_token_accuracy": 0.15694964528083802,
      "num_tokens": 14409063.0,
      "step": 6565
    },
    {
      "entropy": 5.52393217086792,
      "epoch": 0.6520444620881302,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004965480280342839,
      "loss": 5.325,
      "mean_token_accuracy": 0.1627918154001236,
      "num_tokens": 14421056.0,
      "step": 6570
    },
    {
      "entropy": 5.563065719604492,
      "epoch": 0.6525406907502977,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004965418426369419,
      "loss": 5.3247,
      "mean_token_accuracy": 0.16055609956383704,
      "num_tokens": 14430778.0,
      "step": 6575
    },
    {
      "entropy": 5.547006320953369,
      "epoch": 0.6530369194124652,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004965356517458395,
      "loss": 5.3391,
      "mean_token_accuracy": 0.15274855643510818,
      "num_tokens": 14441544.0,
      "step": 6580
    },
    {
      "entropy": 5.579457855224609,
      "epoch": 0.6535331480746328,
      "grad_norm": 0.875,
      "learning_rate": 0.0004965294553611303,
      "loss": 5.4925,
      "mean_token_accuracy": 0.15214899480342864,
      "num_tokens": 14453062.0,
      "step": 6585
    },
    {
      "entropy": 5.602321434020996,
      "epoch": 0.6540293767368003,
      "grad_norm": 0.953125,
      "learning_rate": 0.000496523253482968,
      "loss": 5.4204,
      "mean_token_accuracy": 0.15419695898890495,
      "num_tokens": 14464402.0,
      "step": 6590
    },
    {
      "entropy": 5.635910081863403,
      "epoch": 0.6545256053989679,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965170461115065,
      "loss": 5.5059,
      "mean_token_accuracy": 0.15288595706224442,
      "num_tokens": 14476595.0,
      "step": 6595
    },
    {
      "entropy": 5.700248050689697,
      "epoch": 0.6550218340611353,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004965108332468994,
      "loss": 5.4883,
      "mean_token_accuracy": 0.14744484722614287,
      "num_tokens": 14487298.0,
      "step": 6600
    },
    {
      "entropy": 5.587152910232544,
      "epoch": 0.6555180627233029,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004965046148893011,
      "loss": 5.3789,
      "mean_token_accuracy": 0.16049299538135528,
      "num_tokens": 14499708.0,
      "step": 6605
    },
    {
      "entropy": 5.576786851882934,
      "epoch": 0.6560142913854704,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004964983910388654,
      "loss": 5.4411,
      "mean_token_accuracy": 0.15392221957445146,
      "num_tokens": 14511773.0,
      "step": 6610
    },
    {
      "entropy": 5.547587585449219,
      "epoch": 0.656510520047638,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004964921616957471,
      "loss": 5.3444,
      "mean_token_accuracy": 0.15164763033390044,
      "num_tokens": 14523024.0,
      "step": 6615
    },
    {
      "entropy": 5.617991828918457,
      "epoch": 0.6570067487098055,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004964859268601005,
      "loss": 5.4126,
      "mean_token_accuracy": 0.154441487044096,
      "num_tokens": 14532915.0,
      "step": 6620
    },
    {
      "entropy": 5.598261404037475,
      "epoch": 0.657502977371973,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004964796865320801,
      "loss": 5.4043,
      "mean_token_accuracy": 0.15950134098529817,
      "num_tokens": 14543627.0,
      "step": 6625
    },
    {
      "entropy": 5.525602531433106,
      "epoch": 0.6579992060341405,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004964734407118406,
      "loss": 5.3893,
      "mean_token_accuracy": 0.15673066079616546,
      "num_tokens": 14555181.0,
      "step": 6630
    },
    {
      "entropy": 5.539927768707275,
      "epoch": 0.6584954346963081,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004964671893995372,
      "loss": 5.3771,
      "mean_token_accuracy": 0.1498602196574211,
      "num_tokens": 14565542.0,
      "step": 6635
    },
    {
      "entropy": 5.630816411972046,
      "epoch": 0.6589916633584756,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004964609325953245,
      "loss": 5.4384,
      "mean_token_accuracy": 0.1532747134566307,
      "num_tokens": 14575395.0,
      "step": 6640
    },
    {
      "entropy": 5.629387712478637,
      "epoch": 0.6594878920206431,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004964546702993581,
      "loss": 5.4587,
      "mean_token_accuracy": 0.15217862278223038,
      "num_tokens": 14587206.0,
      "step": 6645
    },
    {
      "entropy": 5.650357103347778,
      "epoch": 0.6599841206828106,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000496448402511793,
      "loss": 5.4758,
      "mean_token_accuracy": 0.15155555158853531,
      "num_tokens": 14599048.0,
      "step": 6650
    },
    {
      "entropy": 5.654848718643189,
      "epoch": 0.6604803493449781,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004964421292327847,
      "loss": 5.4365,
      "mean_token_accuracy": 0.15383079051971435,
      "num_tokens": 14609633.0,
      "step": 6655
    },
    {
      "entropy": 5.5945721626281735,
      "epoch": 0.6609765780071457,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004964358504624887,
      "loss": 5.346,
      "mean_token_accuracy": 0.15625559240579606,
      "num_tokens": 14620373.0,
      "step": 6660
    },
    {
      "entropy": 5.537472248077393,
      "epoch": 0.6614728066693132,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004964295662010608,
      "loss": 5.2967,
      "mean_token_accuracy": 0.16170741766691207,
      "num_tokens": 14630463.0,
      "step": 6665
    },
    {
      "entropy": 5.740504264831543,
      "epoch": 0.6619690353314808,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004964232764486568,
      "loss": 5.5402,
      "mean_token_accuracy": 0.1473265089094639,
      "num_tokens": 14641523.0,
      "step": 6670
    },
    {
      "entropy": 5.589998054504394,
      "epoch": 0.6624652639936482,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004964169812054326,
      "loss": 5.3418,
      "mean_token_accuracy": 0.15613186061382295,
      "num_tokens": 14651812.0,
      "step": 6675
    },
    {
      "entropy": 5.514553642272949,
      "epoch": 0.6629614926558158,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004964106804715444,
      "loss": 5.3556,
      "mean_token_accuracy": 0.15830594450235366,
      "num_tokens": 14661982.0,
      "step": 6680
    },
    {
      "entropy": 5.569979190826416,
      "epoch": 0.6634577213179833,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004964043742471486,
      "loss": 5.3686,
      "mean_token_accuracy": 0.162179996073246,
      "num_tokens": 14671981.0,
      "step": 6685
    },
    {
      "entropy": 5.622475957870483,
      "epoch": 0.6639539499801509,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004963980625324012,
      "loss": 5.3914,
      "mean_token_accuracy": 0.1518598973751068,
      "num_tokens": 14682478.0,
      "step": 6690
    },
    {
      "entropy": 5.633850479125977,
      "epoch": 0.6644501786423184,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000496391745327459,
      "loss": 5.479,
      "mean_token_accuracy": 0.14283154159784317,
      "num_tokens": 14695197.0,
      "step": 6695
    },
    {
      "entropy": 5.586344242095947,
      "epoch": 0.664946407304486,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004963854226324785,
      "loss": 5.3718,
      "mean_token_accuracy": 0.15447912067174913,
      "num_tokens": 14705702.0,
      "step": 6700
    },
    {
      "entropy": 5.602133846282959,
      "epoch": 0.6654426359666534,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004963790944476167,
      "loss": 5.393,
      "mean_token_accuracy": 0.1590353712439537,
      "num_tokens": 14715370.0,
      "step": 6705
    },
    {
      "entropy": 5.636506462097168,
      "epoch": 0.665938864628821,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004963727607730303,
      "loss": 5.4095,
      "mean_token_accuracy": 0.15574731081724166,
      "num_tokens": 14725481.0,
      "step": 6710
    },
    {
      "entropy": 5.5663489818573,
      "epoch": 0.6664350932909885,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004963664216088764,
      "loss": 5.3308,
      "mean_token_accuracy": 0.15330153852701187,
      "num_tokens": 14735250.0,
      "step": 6715
    },
    {
      "entropy": 5.550839567184449,
      "epoch": 0.666931321953156,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963600769553124,
      "loss": 5.3423,
      "mean_token_accuracy": 0.16309174597263337,
      "num_tokens": 14744741.0,
      "step": 6720
    },
    {
      "entropy": 5.648164129257202,
      "epoch": 0.6674275506153236,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004963537268124953,
      "loss": 5.3806,
      "mean_token_accuracy": 0.15598916113376618,
      "num_tokens": 14755865.0,
      "step": 6725
    },
    {
      "entropy": 5.5299581527709964,
      "epoch": 0.667923779277491,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004963473711805829,
      "loss": 5.3308,
      "mean_token_accuracy": 0.1584343433380127,
      "num_tokens": 14766154.0,
      "step": 6730
    },
    {
      "entropy": 5.5833440780639645,
      "epoch": 0.6684200079396586,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004963410100597326,
      "loss": 5.388,
      "mean_token_accuracy": 0.1521584615111351,
      "num_tokens": 14776940.0,
      "step": 6735
    },
    {
      "entropy": 5.560282754898071,
      "epoch": 0.6689162366018261,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004963346434501021,
      "loss": 5.4813,
      "mean_token_accuracy": 0.15100081786513328,
      "num_tokens": 14788367.0,
      "step": 6740
    },
    {
      "entropy": 5.56590051651001,
      "epoch": 0.6694124652639937,
      "grad_norm": 0.875,
      "learning_rate": 0.0004963282713518495,
      "loss": 5.3901,
      "mean_token_accuracy": 0.15709468275308608,
      "num_tokens": 14798751.0,
      "step": 6745
    },
    {
      "entropy": 5.539422416687012,
      "epoch": 0.6699086939261611,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004963218937651326,
      "loss": 5.323,
      "mean_token_accuracy": 0.15921037793159484,
      "num_tokens": 14808748.0,
      "step": 6750
    },
    {
      "entropy": 5.621778106689453,
      "epoch": 0.6704049225883287,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004963155106901098,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1455796867609024,
      "num_tokens": 14820524.0,
      "step": 6755
    },
    {
      "entropy": 5.7212916851043705,
      "epoch": 0.6709011512504962,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004963091221269391,
      "loss": 5.5437,
      "mean_token_accuracy": 0.14160653054714203,
      "num_tokens": 14832485.0,
      "step": 6760
    },
    {
      "entropy": 5.581074953079224,
      "epoch": 0.6713973799126638,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000496302728075779,
      "loss": 5.3353,
      "mean_token_accuracy": 0.1552274703979492,
      "num_tokens": 14843951.0,
      "step": 6765
    },
    {
      "entropy": 5.529450416564941,
      "epoch": 0.6718936085748313,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004962963285367882,
      "loss": 5.4265,
      "mean_token_accuracy": 0.1550913542509079,
      "num_tokens": 14855502.0,
      "step": 6770
    },
    {
      "entropy": 5.558300542831421,
      "epoch": 0.6723898372369989,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004962899235101253,
      "loss": 5.338,
      "mean_token_accuracy": 0.15669150203466414,
      "num_tokens": 14866305.0,
      "step": 6775
    },
    {
      "entropy": 5.592230749130249,
      "epoch": 0.6728860658991663,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004962835129959491,
      "loss": 5.3719,
      "mean_token_accuracy": 0.15086598098278045,
      "num_tokens": 14877222.0,
      "step": 6780
    },
    {
      "entropy": 5.55699896812439,
      "epoch": 0.6733822945613339,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004962770969944186,
      "loss": 5.433,
      "mean_token_accuracy": 0.15394371300935744,
      "num_tokens": 14888099.0,
      "step": 6785
    },
    {
      "entropy": 5.572944831848145,
      "epoch": 0.6738785232235014,
      "grad_norm": 0.9375,
      "learning_rate": 0.000496270675505693,
      "loss": 5.3696,
      "mean_token_accuracy": 0.15748813152313232,
      "num_tokens": 14899042.0,
      "step": 6790
    },
    {
      "entropy": 5.584994602203369,
      "epoch": 0.6743747518856689,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004962642485299314,
      "loss": 5.4143,
      "mean_token_accuracy": 0.15487085282802582,
      "num_tokens": 14908837.0,
      "step": 6795
    },
    {
      "entropy": 5.543886661529541,
      "epoch": 0.6748709805478365,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004962578160672931,
      "loss": 5.3594,
      "mean_token_accuracy": 0.16115720719099044,
      "num_tokens": 14920773.0,
      "step": 6800
    },
    {
      "entropy": 5.5483417987823485,
      "epoch": 0.6753672092100039,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004962513781179379,
      "loss": 5.3534,
      "mean_token_accuracy": 0.15471526086330414,
      "num_tokens": 14930970.0,
      "step": 6805
    },
    {
      "entropy": 5.565474987030029,
      "epoch": 0.6758634378721715,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004962449346820251,
      "loss": 5.4262,
      "mean_token_accuracy": 0.1476544000208378,
      "num_tokens": 14942378.0,
      "step": 6810
    },
    {
      "entropy": 5.558213567733764,
      "epoch": 0.676359666534339,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004962384857597148,
      "loss": 5.4566,
      "mean_token_accuracy": 0.15085810124874116,
      "num_tokens": 14953020.0,
      "step": 6815
    },
    {
      "entropy": 5.653578233718872,
      "epoch": 0.6768558951965066,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004962320313511668,
      "loss": 5.5269,
      "mean_token_accuracy": 0.15361799597740172,
      "num_tokens": 14962340.0,
      "step": 6820
    },
    {
      "entropy": 5.625601720809937,
      "epoch": 0.677352123858674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004962255714565409,
      "loss": 5.4141,
      "mean_token_accuracy": 0.15450359284877777,
      "num_tokens": 14972792.0,
      "step": 6825
    },
    {
      "entropy": 5.571131658554077,
      "epoch": 0.6778483525208416,
      "grad_norm": 0.875,
      "learning_rate": 0.0004962191060759978,
      "loss": 5.3912,
      "mean_token_accuracy": 0.15612896382808686,
      "num_tokens": 14983698.0,
      "step": 6830
    },
    {
      "entropy": 5.613586330413819,
      "epoch": 0.6783445811830091,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004962126352096973,
      "loss": 5.3524,
      "mean_token_accuracy": 0.15863122791051865,
      "num_tokens": 14994583.0,
      "step": 6835
    },
    {
      "entropy": 5.51322569847107,
      "epoch": 0.6788408098451767,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004962061588578004,
      "loss": 5.2983,
      "mean_token_accuracy": 0.15476125478744507,
      "num_tokens": 15005325.0,
      "step": 6840
    },
    {
      "entropy": 5.588961601257324,
      "epoch": 0.6793370385073442,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004961996770204673,
      "loss": 5.3704,
      "mean_token_accuracy": 0.1552986204624176,
      "num_tokens": 15015155.0,
      "step": 6845
    },
    {
      "entropy": 5.578842353820801,
      "epoch": 0.6798332671695118,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004961931896978588,
      "loss": 5.4308,
      "mean_token_accuracy": 0.15067857205867768,
      "num_tokens": 15026529.0,
      "step": 6850
    },
    {
      "entropy": 5.50896635055542,
      "epoch": 0.6803294958316792,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004961866968901359,
      "loss": 5.3618,
      "mean_token_accuracy": 0.15841467380523683,
      "num_tokens": 15037632.0,
      "step": 6855
    },
    {
      "entropy": 5.562885856628418,
      "epoch": 0.6808257244938468,
      "grad_norm": 0.875,
      "learning_rate": 0.0004961801985974595,
      "loss": 5.4317,
      "mean_token_accuracy": 0.15281036049127578,
      "num_tokens": 15050397.0,
      "step": 6860
    },
    {
      "entropy": 5.555966424942016,
      "epoch": 0.6813219531560143,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004961736948199907,
      "loss": 5.3491,
      "mean_token_accuracy": 0.15506883710622787,
      "num_tokens": 15061505.0,
      "step": 6865
    },
    {
      "entropy": 5.660563802719116,
      "epoch": 0.6818181818181818,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004961671855578909,
      "loss": 5.5388,
      "mean_token_accuracy": 0.1459961101412773,
      "num_tokens": 15073752.0,
      "step": 6870
    },
    {
      "entropy": 5.6258409976959225,
      "epoch": 0.6823144104803494,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004961606708113216,
      "loss": 5.4266,
      "mean_token_accuracy": 0.15296203792095184,
      "num_tokens": 15084730.0,
      "step": 6875
    },
    {
      "entropy": 5.583535385131836,
      "epoch": 0.6828106391425168,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004961541505804441,
      "loss": 5.3401,
      "mean_token_accuracy": 0.15618596374988555,
      "num_tokens": 15095811.0,
      "step": 6880
    },
    {
      "entropy": 5.579980182647705,
      "epoch": 0.6833068678046844,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004961476248654202,
      "loss": 5.3299,
      "mean_token_accuracy": 0.16105467826128006,
      "num_tokens": 15106102.0,
      "step": 6885
    },
    {
      "entropy": 5.59256911277771,
      "epoch": 0.6838030964668519,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004961410936664118,
      "loss": 5.4246,
      "mean_token_accuracy": 0.15121853053569795,
      "num_tokens": 15115907.0,
      "step": 6890
    },
    {
      "entropy": 5.580052757263184,
      "epoch": 0.6842993251290195,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004961345569835807,
      "loss": 5.3971,
      "mean_token_accuracy": 0.1558905601501465,
      "num_tokens": 15126788.0,
      "step": 6895
    },
    {
      "entropy": 5.584158658981323,
      "epoch": 0.684795553791187,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004961280148170892,
      "loss": 5.3621,
      "mean_token_accuracy": 0.15638109892606736,
      "num_tokens": 15137814.0,
      "step": 6900
    },
    {
      "entropy": 5.631082773208618,
      "epoch": 0.6852917824533545,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004961214671670994,
      "loss": 5.5891,
      "mean_token_accuracy": 0.1433118350803852,
      "num_tokens": 15148597.0,
      "step": 6905
    },
    {
      "entropy": 5.630214738845825,
      "epoch": 0.685788011115522,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004961149140337737,
      "loss": 5.3957,
      "mean_token_accuracy": 0.1523701250553131,
      "num_tokens": 15158985.0,
      "step": 6910
    },
    {
      "entropy": 5.566757869720459,
      "epoch": 0.6862842397776896,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004961083554172745,
      "loss": 5.3784,
      "mean_token_accuracy": 0.15582308694720268,
      "num_tokens": 15170242.0,
      "step": 6915
    },
    {
      "entropy": 5.5762779712677,
      "epoch": 0.6867804684398571,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004961017913177646,
      "loss": 5.3514,
      "mean_token_accuracy": 0.1617621198296547,
      "num_tokens": 15180353.0,
      "step": 6920
    },
    {
      "entropy": 5.5126060962677,
      "epoch": 0.6872766971020247,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004960952217354066,
      "loss": 5.3043,
      "mean_token_accuracy": 0.16224115192890168,
      "num_tokens": 15191955.0,
      "step": 6925
    },
    {
      "entropy": 5.535926866531372,
      "epoch": 0.6877729257641921,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004960886466703636,
      "loss": 5.3753,
      "mean_token_accuracy": 0.15272082388401031,
      "num_tokens": 15202088.0,
      "step": 6930
    },
    {
      "entropy": 5.515703868865967,
      "epoch": 0.6882691544263597,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004960820661227987,
      "loss": 5.3163,
      "mean_token_accuracy": 0.15685615092515945,
      "num_tokens": 15212291.0,
      "step": 6935
    },
    {
      "entropy": 5.555116033554077,
      "epoch": 0.6887653830885272,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960754800928748,
      "loss": 5.3691,
      "mean_token_accuracy": 0.16065500527620316,
      "num_tokens": 15223315.0,
      "step": 6940
    },
    {
      "entropy": 5.656789302825928,
      "epoch": 0.6892616117506947,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004960688885807553,
      "loss": 5.4829,
      "mean_token_accuracy": 0.15075226426124572,
      "num_tokens": 15233416.0,
      "step": 6945
    },
    {
      "entropy": 5.625695610046387,
      "epoch": 0.6897578404128623,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004960622915866038,
      "loss": 5.3164,
      "mean_token_accuracy": 0.16170334964990615,
      "num_tokens": 15242951.0,
      "step": 6950
    },
    {
      "entropy": 5.542146015167236,
      "epoch": 0.6902540690750297,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004960556891105839,
      "loss": 5.3995,
      "mean_token_accuracy": 0.15278018414974212,
      "num_tokens": 15253145.0,
      "step": 6955
    },
    {
      "entropy": 5.546169090270996,
      "epoch": 0.6907502977371973,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004960490811528591,
      "loss": 5.3452,
      "mean_token_accuracy": 0.16333898156881332,
      "num_tokens": 15263712.0,
      "step": 6960
    },
    {
      "entropy": 5.671264123916626,
      "epoch": 0.6912465263993648,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004960424677135935,
      "loss": 5.4732,
      "mean_token_accuracy": 0.15577918738126756,
      "num_tokens": 15273609.0,
      "step": 6965
    },
    {
      "entropy": 5.5488487720489506,
      "epoch": 0.6917427550615324,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004960358487929511,
      "loss": 5.3341,
      "mean_token_accuracy": 0.16462758779525757,
      "num_tokens": 15283934.0,
      "step": 6970
    },
    {
      "entropy": 5.548047065734863,
      "epoch": 0.6922389837236999,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004960292243910958,
      "loss": 5.3984,
      "mean_token_accuracy": 0.15148687809705735,
      "num_tokens": 15294394.0,
      "step": 6975
    },
    {
      "entropy": 5.582794809341431,
      "epoch": 0.6927352123858674,
      "grad_norm": 0.921875,
      "learning_rate": 0.000496022594508192,
      "loss": 5.3622,
      "mean_token_accuracy": 0.16121286451816558,
      "num_tokens": 15304622.0,
      "step": 6980
    },
    {
      "entropy": 5.567217922210693,
      "epoch": 0.6932314410480349,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004960159591444041,
      "loss": 5.2711,
      "mean_token_accuracy": 0.1600771963596344,
      "num_tokens": 15314094.0,
      "step": 6985
    },
    {
      "entropy": 5.557595539093017,
      "epoch": 0.6937276697102025,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004960093182998968,
      "loss": 5.3916,
      "mean_token_accuracy": 0.16023263037204744,
      "num_tokens": 15324313.0,
      "step": 6990
    },
    {
      "entropy": 5.487686729431152,
      "epoch": 0.69422389837237,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004960026719748346,
      "loss": 5.2706,
      "mean_token_accuracy": 0.17502587735652925,
      "num_tokens": 15335195.0,
      "step": 6995
    },
    {
      "entropy": 5.690118312835693,
      "epoch": 0.6947201270345376,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004959960201693822,
      "loss": 5.5042,
      "mean_token_accuracy": 0.1464548960328102,
      "num_tokens": 15345798.0,
      "step": 7000
    },
    {
      "entropy": 5.568120956420898,
      "epoch": 0.695216355696705,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004959893628837048,
      "loss": 5.3216,
      "mean_token_accuracy": 0.1585929274559021,
      "num_tokens": 15356340.0,
      "step": 7005
    },
    {
      "entropy": 5.606752204895019,
      "epoch": 0.6957125843588726,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004959827001179675,
      "loss": 5.4273,
      "mean_token_accuracy": 0.1500048279762268,
      "num_tokens": 15367795.0,
      "step": 7010
    },
    {
      "entropy": 5.634905433654785,
      "epoch": 0.6962088130210401,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004959760318723353,
      "loss": 5.5592,
      "mean_token_accuracy": 0.14500176906585693,
      "num_tokens": 15379214.0,
      "step": 7015
    },
    {
      "entropy": 5.589799880981445,
      "epoch": 0.6967050416832076,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004959693581469736,
      "loss": 5.3946,
      "mean_token_accuracy": 0.16276417523622513,
      "num_tokens": 15390242.0,
      "step": 7020
    },
    {
      "entropy": 5.543370485305786,
      "epoch": 0.6972012703453752,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004959626789420481,
      "loss": 5.3472,
      "mean_token_accuracy": 0.16189260482788087,
      "num_tokens": 15402600.0,
      "step": 7025
    },
    {
      "entropy": 5.556691026687622,
      "epoch": 0.6976974990075426,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004959559942577243,
      "loss": 5.3659,
      "mean_token_accuracy": 0.15843235999345778,
      "num_tokens": 15412883.0,
      "step": 7030
    },
    {
      "entropy": 5.643964672088623,
      "epoch": 0.6981937276697102,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004959493040941678,
      "loss": 5.4198,
      "mean_token_accuracy": 0.15234392434358596,
      "num_tokens": 15425919.0,
      "step": 7035
    },
    {
      "entropy": 5.552553415298462,
      "epoch": 0.6986899563318777,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004959426084515448,
      "loss": 5.3401,
      "mean_token_accuracy": 0.1577442169189453,
      "num_tokens": 15437604.0,
      "step": 7040
    },
    {
      "entropy": 5.54770336151123,
      "epoch": 0.6991861849940453,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004959359073300212,
      "loss": 5.4063,
      "mean_token_accuracy": 0.15242261290550232,
      "num_tokens": 15449023.0,
      "step": 7045
    },
    {
      "entropy": 5.640282535552979,
      "epoch": 0.6996824136562128,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004959292007297631,
      "loss": 5.4535,
      "mean_token_accuracy": 0.15502861142158508,
      "num_tokens": 15459567.0,
      "step": 7050
    },
    {
      "entropy": 5.591946411132812,
      "epoch": 0.7001786423183803,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000495922488650937,
      "loss": 5.3815,
      "mean_token_accuracy": 0.15652793496847153,
      "num_tokens": 15470950.0,
      "step": 7055
    },
    {
      "entropy": 5.53835244178772,
      "epoch": 0.7006748709805478,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004959157710937092,
      "loss": 5.3718,
      "mean_token_accuracy": 0.15299536138772965,
      "num_tokens": 15481258.0,
      "step": 7060
    },
    {
      "entropy": 5.589594173431396,
      "epoch": 0.7011710996427154,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004959090480582462,
      "loss": 5.3002,
      "mean_token_accuracy": 0.1572806268930435,
      "num_tokens": 15491672.0,
      "step": 7065
    },
    {
      "entropy": 5.533781099319458,
      "epoch": 0.7016673283048829,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004959023195447148,
      "loss": 5.3717,
      "mean_token_accuracy": 0.15430044680833815,
      "num_tokens": 15502342.0,
      "step": 7070
    },
    {
      "entropy": 5.548447132110596,
      "epoch": 0.7021635569670505,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000495895585553282,
      "loss": 5.4055,
      "mean_token_accuracy": 0.1500248782336712,
      "num_tokens": 15513430.0,
      "step": 7075
    },
    {
      "entropy": 5.773319435119629,
      "epoch": 0.7026597856292179,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004958888460841147,
      "loss": 5.5038,
      "mean_token_accuracy": 0.15315764024853706,
      "num_tokens": 15524889.0,
      "step": 7080
    },
    {
      "entropy": 5.56788067817688,
      "epoch": 0.7031560142913855,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004958821011373798,
      "loss": 5.3231,
      "mean_token_accuracy": 0.15642202347517015,
      "num_tokens": 15534998.0,
      "step": 7085
    },
    {
      "entropy": 5.514524364471436,
      "epoch": 0.703652242953553,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004958753507132449,
      "loss": 5.3353,
      "mean_token_accuracy": 0.15758486092090607,
      "num_tokens": 15544752.0,
      "step": 7090
    },
    {
      "entropy": 5.589644813537598,
      "epoch": 0.7041484716157205,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004958685948118772,
      "loss": 5.3486,
      "mean_token_accuracy": 0.1569026455283165,
      "num_tokens": 15555490.0,
      "step": 7095
    },
    {
      "entropy": 5.5551917552948,
      "epoch": 0.7046447002778881,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004958618334334443,
      "loss": 5.2885,
      "mean_token_accuracy": 0.1576731339097023,
      "num_tokens": 15565752.0,
      "step": 7100
    },
    {
      "entropy": 5.55247278213501,
      "epoch": 0.7051409289400555,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004958550665781138,
      "loss": 5.4262,
      "mean_token_accuracy": 0.15204641222953796,
      "num_tokens": 15576368.0,
      "step": 7105
    },
    {
      "entropy": 5.648264694213867,
      "epoch": 0.7056371576022231,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004958482942460536,
      "loss": 5.4218,
      "mean_token_accuracy": 0.15578450560569762,
      "num_tokens": 15587972.0,
      "step": 7110
    },
    {
      "entropy": 5.701238584518433,
      "epoch": 0.7061333862643906,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004958415164374316,
      "loss": 5.4626,
      "mean_token_accuracy": 0.15444232970476152,
      "num_tokens": 15599951.0,
      "step": 7115
    },
    {
      "entropy": 5.559233951568603,
      "epoch": 0.7066296149265582,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004958347331524158,
      "loss": 5.4389,
      "mean_token_accuracy": 0.15119174122810364,
      "num_tokens": 15610183.0,
      "step": 7120
    },
    {
      "entropy": 5.567929315567016,
      "epoch": 0.7071258435887257,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004958279443911745,
      "loss": 5.4394,
      "mean_token_accuracy": 0.15136580318212509,
      "num_tokens": 15621074.0,
      "step": 7125
    },
    {
      "entropy": 5.691843461990357,
      "epoch": 0.7076220722508932,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000495821150153876,
      "loss": 5.449,
      "mean_token_accuracy": 0.15179289430379866,
      "num_tokens": 15632752.0,
      "step": 7130
    },
    {
      "entropy": 5.610294342041016,
      "epoch": 0.7081183009130607,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000495814350440689,
      "loss": 5.4259,
      "mean_token_accuracy": 0.1501299522817135,
      "num_tokens": 15644703.0,
      "step": 7135
    },
    {
      "entropy": 5.569166755676269,
      "epoch": 0.7086145295752283,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004958075452517816,
      "loss": 5.3879,
      "mean_token_accuracy": 0.1546367108821869,
      "num_tokens": 15655278.0,
      "step": 7140
    },
    {
      "entropy": 5.601422882080078,
      "epoch": 0.7091107582373958,
      "grad_norm": 0.875,
      "learning_rate": 0.0004958007345873232,
      "loss": 5.4085,
      "mean_token_accuracy": 0.15654650926589966,
      "num_tokens": 15665998.0,
      "step": 7145
    },
    {
      "entropy": 5.567332220077515,
      "epoch": 0.7096069868995634,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004957939184474821,
      "loss": 5.2947,
      "mean_token_accuracy": 0.15694765597581864,
      "num_tokens": 15676343.0,
      "step": 7150
    },
    {
      "entropy": 5.5898384094238285,
      "epoch": 0.7101032155617308,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004957870968324277,
      "loss": 5.3913,
      "mean_token_accuracy": 0.1532277412712574,
      "num_tokens": 15686520.0,
      "step": 7155
    },
    {
      "entropy": 5.6117500305175785,
      "epoch": 0.7105994442238984,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004957802697423291,
      "loss": 5.5012,
      "mean_token_accuracy": 0.151394285261631,
      "num_tokens": 15698302.0,
      "step": 7160
    },
    {
      "entropy": 5.52632508277893,
      "epoch": 0.7110956728860659,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004957734371773554,
      "loss": 5.2562,
      "mean_token_accuracy": 0.15840691775083543,
      "num_tokens": 15708462.0,
      "step": 7165
    },
    {
      "entropy": 5.590388298034668,
      "epoch": 0.7115919015482334,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957665991376763,
      "loss": 5.378,
      "mean_token_accuracy": 0.16070500314235686,
      "num_tokens": 15719234.0,
      "step": 7170
    },
    {
      "entropy": 5.53080644607544,
      "epoch": 0.712088130210401,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004957597556234613,
      "loss": 5.4013,
      "mean_token_accuracy": 0.15340089201927185,
      "num_tokens": 15732245.0,
      "step": 7175
    },
    {
      "entropy": 5.595386743545532,
      "epoch": 0.7125843588725684,
      "grad_norm": 1.0,
      "learning_rate": 0.0004957529066348799,
      "loss": 5.3819,
      "mean_token_accuracy": 0.16044221967458724,
      "num_tokens": 15742403.0,
      "step": 7180
    },
    {
      "entropy": 5.509410572052002,
      "epoch": 0.713080587534736,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004957460521721022,
      "loss": 5.2863,
      "mean_token_accuracy": 0.1583533227443695,
      "num_tokens": 15752100.0,
      "step": 7185
    },
    {
      "entropy": 5.549450254440307,
      "epoch": 0.7135768161969035,
      "grad_norm": 0.921875,
      "learning_rate": 0.000495739192235298,
      "loss": 5.3787,
      "mean_token_accuracy": 0.154872128367424,
      "num_tokens": 15762315.0,
      "step": 7190
    },
    {
      "entropy": 5.552863550186157,
      "epoch": 0.7140730448590711,
      "grad_norm": 1.0,
      "learning_rate": 0.0004957323268246376,
      "loss": 5.3982,
      "mean_token_accuracy": 0.15654573291540147,
      "num_tokens": 15773841.0,
      "step": 7195
    },
    {
      "entropy": 5.593350791931153,
      "epoch": 0.7145692735212386,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004957254559402911,
      "loss": 5.3425,
      "mean_token_accuracy": 0.15729348957538605,
      "num_tokens": 15785511.0,
      "step": 7200
    },
    {
      "entropy": 5.536798667907715,
      "epoch": 0.7150655021834061,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004957185795824288,
      "loss": 5.3716,
      "mean_token_accuracy": 0.1508410856127739,
      "num_tokens": 15795394.0,
      "step": 7205
    },
    {
      "entropy": 5.525018167495728,
      "epoch": 0.7155617308455736,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004957116977512214,
      "loss": 5.3375,
      "mean_token_accuracy": 0.15722720101475715,
      "num_tokens": 15805894.0,
      "step": 7210
    },
    {
      "entropy": 5.599946451187134,
      "epoch": 0.7160579595077412,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004957048104468395,
      "loss": 5.395,
      "mean_token_accuracy": 0.14500413239002227,
      "num_tokens": 15816595.0,
      "step": 7215
    },
    {
      "entropy": 5.561121034622192,
      "epoch": 0.7165541881699087,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004956979176694539,
      "loss": 5.3367,
      "mean_token_accuracy": 0.16101332306861876,
      "num_tokens": 15827638.0,
      "step": 7220
    },
    {
      "entropy": 5.531207656860351,
      "epoch": 0.7170504168320763,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004956910194192355,
      "loss": 5.4097,
      "mean_token_accuracy": 0.1532890185713768,
      "num_tokens": 15838437.0,
      "step": 7225
    },
    {
      "entropy": 5.690220689773559,
      "epoch": 0.7175466454942437,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004956841156963553,
      "loss": 5.5203,
      "mean_token_accuracy": 0.154303640127182,
      "num_tokens": 15850110.0,
      "step": 7230
    },
    {
      "entropy": 5.58256688117981,
      "epoch": 0.7180428741564113,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004956772065009846,
      "loss": 5.3749,
      "mean_token_accuracy": 0.15135821774601937,
      "num_tokens": 15861786.0,
      "step": 7235
    },
    {
      "entropy": 5.567192411422729,
      "epoch": 0.7185391028185788,
      "grad_norm": 1.0,
      "learning_rate": 0.0004956702918332948,
      "loss": 5.3524,
      "mean_token_accuracy": 0.152687868475914,
      "num_tokens": 15871574.0,
      "step": 7240
    },
    {
      "entropy": 5.578045654296875,
      "epoch": 0.7190353314807463,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004956633716934572,
      "loss": 5.3429,
      "mean_token_accuracy": 0.1614610031247139,
      "num_tokens": 15881944.0,
      "step": 7245
    },
    {
      "entropy": 5.522058582305908,
      "epoch": 0.7195315601429139,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004956564460816433,
      "loss": 5.3467,
      "mean_token_accuracy": 0.1552556797862053,
      "num_tokens": 15892511.0,
      "step": 7250
    },
    {
      "entropy": 5.567608165740967,
      "epoch": 0.7200277888050813,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004956495149980252,
      "loss": 5.3759,
      "mean_token_accuracy": 0.15121624171733855,
      "num_tokens": 15903387.0,
      "step": 7255
    },
    {
      "entropy": 5.598010063171387,
      "epoch": 0.7205240174672489,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004956425784427745,
      "loss": 5.4062,
      "mean_token_accuracy": 0.15305554121732712,
      "num_tokens": 15915117.0,
      "step": 7260
    },
    {
      "entropy": 5.5196246147155765,
      "epoch": 0.7210202461294164,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004956356364160634,
      "loss": 5.3573,
      "mean_token_accuracy": 0.15544371008872987,
      "num_tokens": 15925163.0,
      "step": 7265
    },
    {
      "entropy": 5.449028968811035,
      "epoch": 0.721516474791584,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004956286889180639,
      "loss": 5.2431,
      "mean_token_accuracy": 0.16879490315914153,
      "num_tokens": 15935021.0,
      "step": 7270
    },
    {
      "entropy": 5.512518739700317,
      "epoch": 0.7220127034537515,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004956217359489483,
      "loss": 5.3681,
      "mean_token_accuracy": 0.15262435823678971,
      "num_tokens": 15945416.0,
      "step": 7275
    },
    {
      "entropy": 5.648502159118652,
      "epoch": 0.722508932115919,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000495614777508889,
      "loss": 5.4761,
      "mean_token_accuracy": 0.14991702884435654,
      "num_tokens": 15955388.0,
      "step": 7280
    },
    {
      "entropy": 5.64871301651001,
      "epoch": 0.7230051607780865,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004956078135980586,
      "loss": 5.4853,
      "mean_token_accuracy": 0.15207213908433914,
      "num_tokens": 15966721.0,
      "step": 7285
    },
    {
      "entropy": 5.565513181686401,
      "epoch": 0.7235013894402541,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004956008442166299,
      "loss": 5.4075,
      "mean_token_accuracy": 0.15500583946704866,
      "num_tokens": 15976388.0,
      "step": 7290
    },
    {
      "entropy": 5.521198511123657,
      "epoch": 0.7239976181024216,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004955938693647756,
      "loss": 5.3026,
      "mean_token_accuracy": 0.15797873884439467,
      "num_tokens": 15986632.0,
      "step": 7295
    },
    {
      "entropy": 5.5546427249908445,
      "epoch": 0.7244938467645892,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004955868890426686,
      "loss": 5.3512,
      "mean_token_accuracy": 0.15412055999040603,
      "num_tokens": 15996626.0,
      "step": 7300
    },
    {
      "entropy": 5.508546781539917,
      "epoch": 0.7249900754267566,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004955799032504821,
      "loss": 5.3186,
      "mean_token_accuracy": 0.1552100211381912,
      "num_tokens": 16007055.0,
      "step": 7305
    },
    {
      "entropy": 5.623987674713135,
      "epoch": 0.7254863040889242,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004955729119883893,
      "loss": 5.4046,
      "mean_token_accuracy": 0.15193189680576324,
      "num_tokens": 16019087.0,
      "step": 7310
    },
    {
      "entropy": 5.587416172027588,
      "epoch": 0.7259825327510917,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004955659152565636,
      "loss": 5.3931,
      "mean_token_accuracy": 0.1515917584300041,
      "num_tokens": 16029980.0,
      "step": 7315
    },
    {
      "entropy": 5.5827936172485355,
      "epoch": 0.7264787614132592,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004955589130551784,
      "loss": 5.3693,
      "mean_token_accuracy": 0.15805643126368524,
      "num_tokens": 16040932.0,
      "step": 7320
    },
    {
      "entropy": 5.461764478683472,
      "epoch": 0.7269749900754268,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004955519053844076,
      "loss": 5.3339,
      "mean_token_accuracy": 0.1601037234067917,
      "num_tokens": 16051809.0,
      "step": 7325
    },
    {
      "entropy": 5.5657754898071286,
      "epoch": 0.7274712187375942,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004955448922444246,
      "loss": 5.3872,
      "mean_token_accuracy": 0.15353999137878419,
      "num_tokens": 16062957.0,
      "step": 7330
    },
    {
      "entropy": 5.607474231719971,
      "epoch": 0.7279674473997618,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004955378736354036,
      "loss": 5.4259,
      "mean_token_accuracy": 0.15297536104917525,
      "num_tokens": 16073592.0,
      "step": 7335
    },
    {
      "entropy": 5.541664505004883,
      "epoch": 0.7284636760619293,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004955308495575185,
      "loss": 5.368,
      "mean_token_accuracy": 0.15536474511027337,
      "num_tokens": 16085054.0,
      "step": 7340
    },
    {
      "entropy": 5.595587205886841,
      "epoch": 0.7289599047240969,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004955238200109436,
      "loss": 5.4466,
      "mean_token_accuracy": 0.15401146709918975,
      "num_tokens": 16096936.0,
      "step": 7345
    },
    {
      "entropy": 5.619851350784302,
      "epoch": 0.7294561333862644,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004955167849958531,
      "loss": 5.468,
      "mean_token_accuracy": 0.15150101929903032,
      "num_tokens": 16107673.0,
      "step": 7350
    },
    {
      "entropy": 5.576428174972534,
      "epoch": 0.729952362048432,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004955097445124215,
      "loss": 5.3278,
      "mean_token_accuracy": 0.16196224093437195,
      "num_tokens": 16117941.0,
      "step": 7355
    },
    {
      "entropy": 5.561256551742554,
      "epoch": 0.7304485907105994,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004955026985608235,
      "loss": 5.295,
      "mean_token_accuracy": 0.16231960356235503,
      "num_tokens": 16128737.0,
      "step": 7360
    },
    {
      "entropy": 5.662528657913208,
      "epoch": 0.730944819372767,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004954956471412336,
      "loss": 5.4283,
      "mean_token_accuracy": 0.15473217517137527,
      "num_tokens": 16139599.0,
      "step": 7365
    },
    {
      "entropy": 5.456266450881958,
      "epoch": 0.7314410480349345,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004954885902538269,
      "loss": 5.2789,
      "mean_token_accuracy": 0.1517025738954544,
      "num_tokens": 16150796.0,
      "step": 7370
    },
    {
      "entropy": 5.545576524734497,
      "epoch": 0.7319372766971021,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004954815278987781,
      "loss": 5.3452,
      "mean_token_accuracy": 0.15946361273527146,
      "num_tokens": 16161236.0,
      "step": 7375
    },
    {
      "entropy": 5.6335042953491214,
      "epoch": 0.7324335053592695,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004954744600762627,
      "loss": 5.4484,
      "mean_token_accuracy": 0.15341491848230362,
      "num_tokens": 16172210.0,
      "step": 7380
    },
    {
      "entropy": 5.5065947532653805,
      "epoch": 0.7329297340214371,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004954673867864555,
      "loss": 5.3199,
      "mean_token_accuracy": 0.16296881139278413,
      "num_tokens": 16182950.0,
      "step": 7385
    },
    {
      "entropy": 5.499658250808716,
      "epoch": 0.7334259626836046,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004954603080295323,
      "loss": 5.2341,
      "mean_token_accuracy": 0.16565425992012023,
      "num_tokens": 16193971.0,
      "step": 7390
    },
    {
      "entropy": 5.560158967971802,
      "epoch": 0.7339221913457721,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004954532238056686,
      "loss": 5.3644,
      "mean_token_accuracy": 0.15560306906700133,
      "num_tokens": 16205774.0,
      "step": 7395
    },
    {
      "entropy": 5.539269638061524,
      "epoch": 0.7344184200079397,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004954461341150399,
      "loss": 5.3567,
      "mean_token_accuracy": 0.15304027497768402,
      "num_tokens": 16216626.0,
      "step": 7400
    },
    {
      "entropy": 5.529664468765259,
      "epoch": 0.7349146486701071,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004954390389578222,
      "loss": 5.4158,
      "mean_token_accuracy": 0.1536064475774765,
      "num_tokens": 16227748.0,
      "step": 7405
    },
    {
      "entropy": 5.623319101333618,
      "epoch": 0.7354108773322747,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004954319383341912,
      "loss": 5.4101,
      "mean_token_accuracy": 0.1580138735473156,
      "num_tokens": 16240390.0,
      "step": 7410
    },
    {
      "entropy": 5.578856325149536,
      "epoch": 0.7359071059944422,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004954248322443232,
      "loss": 5.3393,
      "mean_token_accuracy": 0.1636115148663521,
      "num_tokens": 16250563.0,
      "step": 7415
    },
    {
      "entropy": 5.562770986557007,
      "epoch": 0.7364033346566098,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004954177206883943,
      "loss": 5.3993,
      "mean_token_accuracy": 0.14981402605772018,
      "num_tokens": 16260722.0,
      "step": 7420
    },
    {
      "entropy": 5.681869029998779,
      "epoch": 0.7368995633187773,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954106036665809,
      "loss": 5.4131,
      "mean_token_accuracy": 0.15602878108620644,
      "num_tokens": 16270258.0,
      "step": 7425
    },
    {
      "entropy": 5.556429052352906,
      "epoch": 0.7373957919809448,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004954034811790594,
      "loss": 5.2887,
      "mean_token_accuracy": 0.15933127403259278,
      "num_tokens": 16279504.0,
      "step": 7430
    },
    {
      "entropy": 5.492929124832154,
      "epoch": 0.7378920206431123,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004953963532260067,
      "loss": 5.3013,
      "mean_token_accuracy": 0.16148138046264648,
      "num_tokens": 16291327.0,
      "step": 7435
    },
    {
      "entropy": 5.554673480987549,
      "epoch": 0.7383882493052799,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004953892198075991,
      "loss": 5.4413,
      "mean_token_accuracy": 0.1550499364733696,
      "num_tokens": 16302778.0,
      "step": 7440
    },
    {
      "entropy": 5.6280824661254885,
      "epoch": 0.7388844779674474,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004953820809240138,
      "loss": 5.383,
      "mean_token_accuracy": 0.14624368250370026,
      "num_tokens": 16314373.0,
      "step": 7445
    },
    {
      "entropy": 5.553422307968139,
      "epoch": 0.739380706629615,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004953749365754279,
      "loss": 5.3488,
      "mean_token_accuracy": 0.16128434240818024,
      "num_tokens": 16324518.0,
      "step": 7450
    },
    {
      "entropy": 5.535141515731811,
      "epoch": 0.7398769352917824,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004953677867620184,
      "loss": 5.2936,
      "mean_token_accuracy": 0.16068151593208313,
      "num_tokens": 16335657.0,
      "step": 7455
    },
    {
      "entropy": 5.587289524078369,
      "epoch": 0.74037316395395,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004953606314839626,
      "loss": 5.4918,
      "mean_token_accuracy": 0.1496763989329338,
      "num_tokens": 16346497.0,
      "step": 7460
    },
    {
      "entropy": 5.520878314971924,
      "epoch": 0.7408693926161175,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000495353470741438,
      "loss": 5.3644,
      "mean_token_accuracy": 0.15484014749526978,
      "num_tokens": 16358175.0,
      "step": 7465
    },
    {
      "entropy": 5.629048538208008,
      "epoch": 0.741365621278285,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000495346304534622,
      "loss": 5.4611,
      "mean_token_accuracy": 0.15078660398721694,
      "num_tokens": 16369803.0,
      "step": 7470
    },
    {
      "entropy": 5.515393447875977,
      "epoch": 0.7418618499404526,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004953391328636926,
      "loss": 5.1748,
      "mean_token_accuracy": 0.16214750409126283,
      "num_tokens": 16380535.0,
      "step": 7475
    },
    {
      "entropy": 5.490961980819702,
      "epoch": 0.74235807860262,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004953319557288275,
      "loss": 5.3495,
      "mean_token_accuracy": 0.15547017753124237,
      "num_tokens": 16390518.0,
      "step": 7480
    },
    {
      "entropy": 5.531258535385132,
      "epoch": 0.7428543072647876,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004953247731302046,
      "loss": 5.3727,
      "mean_token_accuracy": 0.1484750598669052,
      "num_tokens": 16400627.0,
      "step": 7485
    },
    {
      "entropy": 5.573671674728393,
      "epoch": 0.7433505359269551,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004953175850680022,
      "loss": 5.436,
      "mean_token_accuracy": 0.1594882406294346,
      "num_tokens": 16411136.0,
      "step": 7490
    },
    {
      "entropy": 5.659179639816284,
      "epoch": 0.7438467645891227,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004953103915423983,
      "loss": 5.3683,
      "mean_token_accuracy": 0.15788947343826293,
      "num_tokens": 16422830.0,
      "step": 7495
    },
    {
      "entropy": 5.513751602172851,
      "epoch": 0.7443429932512902,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004953031925535715,
      "loss": 5.3206,
      "mean_token_accuracy": 0.162714684009552,
      "num_tokens": 16434158.0,
      "step": 7500
    },
    {
      "entropy": 5.512455320358276,
      "epoch": 0.7448392219134577,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004952959881017003,
      "loss": 5.3389,
      "mean_token_accuracy": 0.1615119531750679,
      "num_tokens": 16444562.0,
      "step": 7505
    },
    {
      "entropy": 5.5862266540527346,
      "epoch": 0.7453354505756252,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004952887781869631,
      "loss": 5.3818,
      "mean_token_accuracy": 0.15387180745601653,
      "num_tokens": 16456353.0,
      "step": 7510
    },
    {
      "entropy": 5.5833018779754635,
      "epoch": 0.7458316792377928,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004952815628095391,
      "loss": 5.3976,
      "mean_token_accuracy": 0.14866131767630578,
      "num_tokens": 16465756.0,
      "step": 7515
    },
    {
      "entropy": 5.537196159362793,
      "epoch": 0.7463279078999603,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004952743419696069,
      "loss": 5.3311,
      "mean_token_accuracy": 0.1572333201766014,
      "num_tokens": 16477562.0,
      "step": 7520
    },
    {
      "entropy": 5.603846645355224,
      "epoch": 0.7468241365621279,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004952671156673456,
      "loss": 5.3832,
      "mean_token_accuracy": 0.152971251308918,
      "num_tokens": 16488134.0,
      "step": 7525
    },
    {
      "entropy": 5.591313695907592,
      "epoch": 0.7473203652242953,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004952598839029345,
      "loss": 5.3663,
      "mean_token_accuracy": 0.15624374002218247,
      "num_tokens": 16499146.0,
      "step": 7530
    },
    {
      "entropy": 5.573712968826294,
      "epoch": 0.7478165938864629,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004952526466765529,
      "loss": 5.4145,
      "mean_token_accuracy": 0.15605186074972152,
      "num_tokens": 16510334.0,
      "step": 7535
    },
    {
      "entropy": 5.56377158164978,
      "epoch": 0.7483128225486304,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004952454039883804,
      "loss": 5.2854,
      "mean_token_accuracy": 0.15974623411893846,
      "num_tokens": 16520574.0,
      "step": 7540
    },
    {
      "entropy": 5.597139978408814,
      "epoch": 0.7488090512107979,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004952381558385963,
      "loss": 5.4305,
      "mean_token_accuracy": 0.15135585218667985,
      "num_tokens": 16532246.0,
      "step": 7545
    },
    {
      "entropy": 5.704423427581787,
      "epoch": 0.7493052798729655,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004952309022273805,
      "loss": 5.4,
      "mean_token_accuracy": 0.15351941138505937,
      "num_tokens": 16541756.0,
      "step": 7550
    },
    {
      "entropy": 5.586373615264892,
      "epoch": 0.7498015085351329,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004952236431549129,
      "loss": 5.4054,
      "mean_token_accuracy": 0.1476966083049774,
      "num_tokens": 16551061.0,
      "step": 7555
    },
    {
      "entropy": 5.5230710983276365,
      "epoch": 0.7502977371973005,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004952163786213735,
      "loss": 5.352,
      "mean_token_accuracy": 0.15240754783153534,
      "num_tokens": 16561155.0,
      "step": 7560
    },
    {
      "entropy": 5.549474906921387,
      "epoch": 0.750793965859468,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004952091086269425,
      "loss": 5.3117,
      "mean_token_accuracy": 0.14989598616957664,
      "num_tokens": 16571843.0,
      "step": 7565
    },
    {
      "entropy": 5.5611807346344,
      "epoch": 0.7512901945216356,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004952018331718001,
      "loss": 5.3909,
      "mean_token_accuracy": 0.15381356999278067,
      "num_tokens": 16583395.0,
      "step": 7570
    },
    {
      "entropy": 5.633284044265747,
      "epoch": 0.7517864231838031,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004951945522561268,
      "loss": 5.4386,
      "mean_token_accuracy": 0.15344821512699128,
      "num_tokens": 16594545.0,
      "step": 7575
    },
    {
      "entropy": 5.657517290115356,
      "epoch": 0.7522826518459707,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004951872658801029,
      "loss": 5.4965,
      "mean_token_accuracy": 0.15352494940161704,
      "num_tokens": 16605901.0,
      "step": 7580
    },
    {
      "entropy": 5.569665765762329,
      "epoch": 0.7527788805081381,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004951799740439093,
      "loss": 5.4758,
      "mean_token_accuracy": 0.14774393886327744,
      "num_tokens": 16616783.0,
      "step": 7585
    },
    {
      "entropy": 5.565461921691894,
      "epoch": 0.7532751091703057,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000495172676747727,
      "loss": 5.3861,
      "mean_token_accuracy": 0.15814861059188842,
      "num_tokens": 16628655.0,
      "step": 7590
    },
    {
      "entropy": 5.602010869979859,
      "epoch": 0.7537713378324732,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004951653739917366,
      "loss": 5.3932,
      "mean_token_accuracy": 0.1518102467060089,
      "num_tokens": 16639403.0,
      "step": 7595
    },
    {
      "entropy": 5.6291468143463135,
      "epoch": 0.7542675664946408,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004951580657761192,
      "loss": 5.4034,
      "mean_token_accuracy": 0.15876073837280275,
      "num_tokens": 16649937.0,
      "step": 7600
    },
    {
      "entropy": 5.473993968963623,
      "epoch": 0.7547637951568082,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004951507521010563,
      "loss": 5.2965,
      "mean_token_accuracy": 0.15884831845760344,
      "num_tokens": 16661466.0,
      "step": 7605
    },
    {
      "entropy": 5.507606792449951,
      "epoch": 0.7552600238189758,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004951434329667293,
      "loss": 5.3134,
      "mean_token_accuracy": 0.15895138829946517,
      "num_tokens": 16673362.0,
      "step": 7610
    },
    {
      "entropy": 5.548601722717285,
      "epoch": 0.7557562524811433,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004951361083733192,
      "loss": 5.3543,
      "mean_token_accuracy": 0.1521439790725708,
      "num_tokens": 16685555.0,
      "step": 7615
    },
    {
      "entropy": 5.573924922943116,
      "epoch": 0.7562524811433108,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004951287783210082,
      "loss": 5.336,
      "mean_token_accuracy": 0.15216623693704606,
      "num_tokens": 16696006.0,
      "step": 7620
    },
    {
      "entropy": 5.557129573822022,
      "epoch": 0.7567487098054784,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004951214428099778,
      "loss": 5.335,
      "mean_token_accuracy": 0.15660312324762343,
      "num_tokens": 16706784.0,
      "step": 7625
    },
    {
      "entropy": 5.508156061172485,
      "epoch": 0.7572449384676458,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004951141018404099,
      "loss": 5.3166,
      "mean_token_accuracy": 0.1631629690527916,
      "num_tokens": 16718492.0,
      "step": 7630
    },
    {
      "entropy": 5.532926273345947,
      "epoch": 0.7577411671298134,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004951067554124865,
      "loss": 5.3799,
      "mean_token_accuracy": 0.15535615980625153,
      "num_tokens": 16729711.0,
      "step": 7635
    },
    {
      "entropy": 5.538987731933593,
      "epoch": 0.7582373957919809,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049509940352639,
      "loss": 5.3595,
      "mean_token_accuracy": 0.15555189549922943,
      "num_tokens": 16739913.0,
      "step": 7640
    },
    {
      "entropy": 5.638753747940063,
      "epoch": 0.7587336244541485,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004950920461823024,
      "loss": 5.4033,
      "mean_token_accuracy": 0.1522645890712738,
      "num_tokens": 16752273.0,
      "step": 7645
    },
    {
      "entropy": 5.5679864406585695,
      "epoch": 0.759229853116316,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004950846833804064,
      "loss": 5.34,
      "mean_token_accuracy": 0.15633462369441986,
      "num_tokens": 16762050.0,
      "step": 7650
    },
    {
      "entropy": 5.5330182075500485,
      "epoch": 0.7597260817784836,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004950773151208845,
      "loss": 5.4032,
      "mean_token_accuracy": 0.15630503445863725,
      "num_tokens": 16772376.0,
      "step": 7655
    },
    {
      "entropy": 5.546814680099487,
      "epoch": 0.760222310440651,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004950699414039194,
      "loss": 5.3648,
      "mean_token_accuracy": 0.15277177542448045,
      "num_tokens": 16783307.0,
      "step": 7660
    },
    {
      "entropy": 5.574085474014282,
      "epoch": 0.7607185391028186,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000495062562229694,
      "loss": 5.2402,
      "mean_token_accuracy": 0.16051952093839644,
      "num_tokens": 16793582.0,
      "step": 7665
    },
    {
      "entropy": 5.505533981323242,
      "epoch": 0.7612147677649861,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004950551775983912,
      "loss": 5.3639,
      "mean_token_accuracy": 0.1557060956954956,
      "num_tokens": 16805160.0,
      "step": 7670
    },
    {
      "entropy": 5.5394867897033695,
      "epoch": 0.7617109964271537,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004950477875101942,
      "loss": 5.4194,
      "mean_token_accuracy": 0.15286368802189826,
      "num_tokens": 16816108.0,
      "step": 7675
    },
    {
      "entropy": 5.521533107757568,
      "epoch": 0.7622072250893211,
      "grad_norm": 0.875,
      "learning_rate": 0.0004950403919652863,
      "loss": 5.2943,
      "mean_token_accuracy": 0.15909659415483474,
      "num_tokens": 16828126.0,
      "step": 7680
    },
    {
      "entropy": 5.584299373626709,
      "epoch": 0.7627034537514887,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004950329909638507,
      "loss": 5.3624,
      "mean_token_accuracy": 0.155840203166008,
      "num_tokens": 16838186.0,
      "step": 7685
    },
    {
      "entropy": 5.565781497955323,
      "epoch": 0.7631996824136562,
      "grad_norm": 0.80078125,
      "learning_rate": 0.000495025584506071,
      "loss": 5.3038,
      "mean_token_accuracy": 0.15668020844459535,
      "num_tokens": 16848588.0,
      "step": 7690
    },
    {
      "entropy": 5.584897136688232,
      "epoch": 0.7636959110758237,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004950181725921311,
      "loss": 5.4198,
      "mean_token_accuracy": 0.1540161445736885,
      "num_tokens": 16860269.0,
      "step": 7695
    },
    {
      "entropy": 5.563757610321045,
      "epoch": 0.7641921397379913,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004950107552222145,
      "loss": 5.4135,
      "mean_token_accuracy": 0.1542084261775017,
      "num_tokens": 16872221.0,
      "step": 7700
    },
    {
      "entropy": 5.539776229858399,
      "epoch": 0.7646883684001587,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004950033323965052,
      "loss": 5.2995,
      "mean_token_accuracy": 0.1555795654654503,
      "num_tokens": 16882588.0,
      "step": 7705
    },
    {
      "entropy": 5.605736160278321,
      "epoch": 0.7651845970623263,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004949959041151875,
      "loss": 5.3689,
      "mean_token_accuracy": 0.15667574852705002,
      "num_tokens": 16894278.0,
      "step": 7710
    },
    {
      "entropy": 5.567713689804077,
      "epoch": 0.7656808257244938,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004949884703784453,
      "loss": 5.3413,
      "mean_token_accuracy": 0.15670668408274652,
      "num_tokens": 16905448.0,
      "step": 7715
    },
    {
      "entropy": 5.53463191986084,
      "epoch": 0.7661770543866614,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000494981031186463,
      "loss": 5.3692,
      "mean_token_accuracy": 0.15420209094882012,
      "num_tokens": 16914960.0,
      "step": 7720
    },
    {
      "entropy": 5.57317419052124,
      "epoch": 0.7666732830488289,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004949735865394253,
      "loss": 5.4368,
      "mean_token_accuracy": 0.15323926880955696,
      "num_tokens": 16926839.0,
      "step": 7725
    },
    {
      "entropy": 5.56858344078064,
      "epoch": 0.7671695117109965,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004949661364375167,
      "loss": 5.312,
      "mean_token_accuracy": 0.16317027509212495,
      "num_tokens": 16938553.0,
      "step": 7730
    },
    {
      "entropy": 5.59964017868042,
      "epoch": 0.7676657403731639,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004949586808809217,
      "loss": 5.3896,
      "mean_token_accuracy": 0.15333618819713593,
      "num_tokens": 16947987.0,
      "step": 7735
    },
    {
      "entropy": 5.489641618728638,
      "epoch": 0.7681619690353315,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004949512198698256,
      "loss": 5.3169,
      "mean_token_accuracy": 0.15825526118278505,
      "num_tokens": 16959359.0,
      "step": 7740
    },
    {
      "entropy": 5.5458088397979735,
      "epoch": 0.768658197697499,
      "grad_norm": 0.875,
      "learning_rate": 0.000494943753404413,
      "loss": 5.3621,
      "mean_token_accuracy": 0.15992436558008194,
      "num_tokens": 16970731.0,
      "step": 7745
    },
    {
      "entropy": 5.507170724868774,
      "epoch": 0.7691544263596666,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004949362814848693,
      "loss": 5.3168,
      "mean_token_accuracy": 0.16060893684625627,
      "num_tokens": 16981868.0,
      "step": 7750
    },
    {
      "entropy": 5.616957569122315,
      "epoch": 0.769650655021834,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004949288041113799,
      "loss": 5.3748,
      "mean_token_accuracy": 0.1563388839364052,
      "num_tokens": 16994639.0,
      "step": 7755
    },
    {
      "entropy": 5.518722724914551,
      "epoch": 0.7701468836840016,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004949213212841299,
      "loss": 5.3671,
      "mean_token_accuracy": 0.15757046043872833,
      "num_tokens": 17005067.0,
      "step": 7760
    },
    {
      "entropy": 5.533456134796142,
      "epoch": 0.7706431123461691,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000494913833003305,
      "loss": 5.2862,
      "mean_token_accuracy": 0.16234548538923263,
      "num_tokens": 17015102.0,
      "step": 7765
    },
    {
      "entropy": 5.556436967849732,
      "epoch": 0.7711393410083366,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000494906339269091,
      "loss": 5.3135,
      "mean_token_accuracy": 0.1586420491337776,
      "num_tokens": 17026290.0,
      "step": 7770
    },
    {
      "entropy": 5.547585296630859,
      "epoch": 0.7716355696705042,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004948988400816736,
      "loss": 5.2758,
      "mean_token_accuracy": 0.1543065443634987,
      "num_tokens": 17035445.0,
      "step": 7775
    },
    {
      "entropy": 5.577733421325684,
      "epoch": 0.7721317983326716,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948913354412388,
      "loss": 5.3868,
      "mean_token_accuracy": 0.15804281383752822,
      "num_tokens": 17046515.0,
      "step": 7780
    },
    {
      "entropy": 5.586125135421753,
      "epoch": 0.7726280269948392,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004948838253479727,
      "loss": 5.3576,
      "mean_token_accuracy": 0.15887074917554855,
      "num_tokens": 17056988.0,
      "step": 7785
    },
    {
      "entropy": 5.618466567993164,
      "epoch": 0.7731242556570067,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004948763098020616,
      "loss": 5.4159,
      "mean_token_accuracy": 0.14824309647083284,
      "num_tokens": 17068304.0,
      "step": 7790
    },
    {
      "entropy": 5.5217780590057375,
      "epoch": 0.7736204843191743,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948687888036918,
      "loss": 5.328,
      "mean_token_accuracy": 0.1570253327488899,
      "num_tokens": 17078405.0,
      "step": 7795
    },
    {
      "entropy": 5.502141380310059,
      "epoch": 0.7741167129813418,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948612623530497,
      "loss": 5.2224,
      "mean_token_accuracy": 0.159689000248909,
      "num_tokens": 17087839.0,
      "step": 7800
    },
    {
      "entropy": 5.461894512176514,
      "epoch": 0.7746129416435094,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004948537304503221,
      "loss": 5.3239,
      "mean_token_accuracy": 0.15929722636938096,
      "num_tokens": 17099386.0,
      "step": 7805
    },
    {
      "entropy": 5.602202606201172,
      "epoch": 0.7751091703056768,
      "grad_norm": 0.875,
      "learning_rate": 0.0004948461930956959,
      "loss": 5.399,
      "mean_token_accuracy": 0.1521478608250618,
      "num_tokens": 17111754.0,
      "step": 7810
    },
    {
      "entropy": 5.561999034881592,
      "epoch": 0.7756053989678444,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004948386502893577,
      "loss": 5.2793,
      "mean_token_accuracy": 0.162553308904171,
      "num_tokens": 17122882.0,
      "step": 7815
    },
    {
      "entropy": 5.533847236633301,
      "epoch": 0.7761016276300119,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004948311020314946,
      "loss": 5.3924,
      "mean_token_accuracy": 0.1520935371518135,
      "num_tokens": 17133986.0,
      "step": 7820
    },
    {
      "entropy": 5.56985559463501,
      "epoch": 0.7765978562921795,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004948235483222941,
      "loss": 5.3505,
      "mean_token_accuracy": 0.16134029030799865,
      "num_tokens": 17144784.0,
      "step": 7825
    },
    {
      "entropy": 5.60482759475708,
      "epoch": 0.777094084954347,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004948159891619431,
      "loss": 5.4663,
      "mean_token_accuracy": 0.14836366921663285,
      "num_tokens": 17156328.0,
      "step": 7830
    },
    {
      "entropy": 5.627248239517212,
      "epoch": 0.7775903136165145,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004948084245506293,
      "loss": 5.2866,
      "mean_token_accuracy": 0.15721032172441482,
      "num_tokens": 17168062.0,
      "step": 7835
    },
    {
      "entropy": 5.542794609069825,
      "epoch": 0.778086542278682,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00049480085448854,
      "loss": 5.3151,
      "mean_token_accuracy": 0.15671407729387282,
      "num_tokens": 17178872.0,
      "step": 7840
    },
    {
      "entropy": 5.547116708755493,
      "epoch": 0.7785827709408495,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004947932789758635,
      "loss": 5.3648,
      "mean_token_accuracy": 0.15643517822027206,
      "num_tokens": 17189445.0,
      "step": 7845
    },
    {
      "entropy": 5.535507345199585,
      "epoch": 0.7790789996030171,
      "grad_norm": 1.015625,
      "learning_rate": 0.000494785698012787,
      "loss": 5.3167,
      "mean_token_accuracy": 0.15999527126550675,
      "num_tokens": 17199454.0,
      "step": 7850
    },
    {
      "entropy": 5.541446352005005,
      "epoch": 0.7795752282651845,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004947781115994989,
      "loss": 5.3837,
      "mean_token_accuracy": 0.15075388327240943,
      "num_tokens": 17210729.0,
      "step": 7855
    },
    {
      "entropy": 5.557825946807862,
      "epoch": 0.7800714569273521,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004947705197361872,
      "loss": 5.4357,
      "mean_token_accuracy": 0.15301606208086013,
      "num_tokens": 17222961.0,
      "step": 7860
    },
    {
      "entropy": 5.54502305984497,
      "epoch": 0.7805676855895196,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004947629224230401,
      "loss": 5.3163,
      "mean_token_accuracy": 0.1613306075334549,
      "num_tokens": 17234026.0,
      "step": 7865
    },
    {
      "entropy": 5.628870868682862,
      "epoch": 0.7810639142516872,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004947553196602461,
      "loss": 5.4426,
      "mean_token_accuracy": 0.15443769246339797,
      "num_tokens": 17244643.0,
      "step": 7870
    },
    {
      "entropy": 5.638962697982788,
      "epoch": 0.7815601429138547,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004947477114479939,
      "loss": 5.3642,
      "mean_token_accuracy": 0.16114916205406188,
      "num_tokens": 17255382.0,
      "step": 7875
    },
    {
      "entropy": 5.512335777282715,
      "epoch": 0.7820563715760223,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004947400977864718,
      "loss": 5.3355,
      "mean_token_accuracy": 0.15462656915187836,
      "num_tokens": 17265860.0,
      "step": 7880
    },
    {
      "entropy": 5.566488218307495,
      "epoch": 0.7825526002381897,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004947324786758688,
      "loss": 5.4347,
      "mean_token_accuracy": 0.15429544374346732,
      "num_tokens": 17277570.0,
      "step": 7885
    },
    {
      "entropy": 5.661877918243408,
      "epoch": 0.7830488289003573,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004947248541163738,
      "loss": 5.4724,
      "mean_token_accuracy": 0.14807592257857322,
      "num_tokens": 17288684.0,
      "step": 7890
    },
    {
      "entropy": 5.573632764816284,
      "epoch": 0.7835450575625248,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004947172241081759,
      "loss": 5.3097,
      "mean_token_accuracy": 0.15547684729099273,
      "num_tokens": 17299541.0,
      "step": 7895
    },
    {
      "entropy": 5.56752610206604,
      "epoch": 0.7840412862246924,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004947095886514643,
      "loss": 5.2732,
      "mean_token_accuracy": 0.16467752754688264,
      "num_tokens": 17310243.0,
      "step": 7900
    },
    {
      "entropy": 5.575685262680054,
      "epoch": 0.7845375148868599,
      "grad_norm": 1.0,
      "learning_rate": 0.0004947019477464283,
      "loss": 5.3846,
      "mean_token_accuracy": 0.15891839414834977,
      "num_tokens": 17321167.0,
      "step": 7905
    },
    {
      "entropy": 5.497011375427246,
      "epoch": 0.7850337435490274,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004946943013932574,
      "loss": 5.3484,
      "mean_token_accuracy": 0.1585032068192959,
      "num_tokens": 17331076.0,
      "step": 7910
    },
    {
      "entropy": 5.605291795730591,
      "epoch": 0.7855299722111949,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004946866495921413,
      "loss": 5.3346,
      "mean_token_accuracy": 0.15110956281423568,
      "num_tokens": 17341337.0,
      "step": 7915
    },
    {
      "entropy": 5.592031002044678,
      "epoch": 0.7860262008733624,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004946789923432695,
      "loss": 5.3579,
      "mean_token_accuracy": 0.1544201672077179,
      "num_tokens": 17353055.0,
      "step": 7920
    },
    {
      "entropy": 5.4604369640350345,
      "epoch": 0.78652242953553,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004946713296468322,
      "loss": 5.2597,
      "mean_token_accuracy": 0.1598164975643158,
      "num_tokens": 17364442.0,
      "step": 7925
    },
    {
      "entropy": 5.5707447052001955,
      "epoch": 0.7870186581976975,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946636615030192,
      "loss": 5.3883,
      "mean_token_accuracy": 0.15719339847564698,
      "num_tokens": 17376237.0,
      "step": 7930
    },
    {
      "entropy": 5.5942220211029055,
      "epoch": 0.787514886859865,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004946559879120208,
      "loss": 5.31,
      "mean_token_accuracy": 0.1578843578696251,
      "num_tokens": 17386188.0,
      "step": 7935
    },
    {
      "entropy": 5.489248991012573,
      "epoch": 0.7880111155220325,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004946483088740273,
      "loss": 5.3107,
      "mean_token_accuracy": 0.16017146706581115,
      "num_tokens": 17397058.0,
      "step": 7940
    },
    {
      "entropy": 5.601967000961304,
      "epoch": 0.7885073441842001,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946406243892289,
      "loss": 5.3578,
      "mean_token_accuracy": 0.15778404772281646,
      "num_tokens": 17408191.0,
      "step": 7945
    },
    {
      "entropy": 5.567347288131714,
      "epoch": 0.7890035728463676,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004946329344578163,
      "loss": 5.322,
      "mean_token_accuracy": 0.15752597898244858,
      "num_tokens": 17418472.0,
      "step": 7950
    },
    {
      "entropy": 5.563692188262939,
      "epoch": 0.7894998015085352,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004946252390799803,
      "loss": 5.2642,
      "mean_token_accuracy": 0.15932658165693284,
      "num_tokens": 17429556.0,
      "step": 7955
    },
    {
      "entropy": 5.586987924575806,
      "epoch": 0.7899960301707026,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004946175382559116,
      "loss": 5.4373,
      "mean_token_accuracy": 0.14806602373719216,
      "num_tokens": 17440950.0,
      "step": 7960
    },
    {
      "entropy": 5.503143644332885,
      "epoch": 0.7904922588328702,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000494609831985801,
      "loss": 5.3143,
      "mean_token_accuracy": 0.16252274811267853,
      "num_tokens": 17453724.0,
      "step": 7965
    },
    {
      "entropy": 5.572233009338379,
      "epoch": 0.7909884874950377,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00049460212026984,
      "loss": 5.3442,
      "mean_token_accuracy": 0.1571431815624237,
      "num_tokens": 17464623.0,
      "step": 7970
    },
    {
      "entropy": 5.528927183151245,
      "epoch": 0.7914847161572053,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004945944031082195,
      "loss": 5.3417,
      "mean_token_accuracy": 0.15848769694566728,
      "num_tokens": 17474919.0,
      "step": 7975
    },
    {
      "entropy": 5.589403867721558,
      "epoch": 0.7919809448193728,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000494586680501131,
      "loss": 5.3415,
      "mean_token_accuracy": 0.1526467278599739,
      "num_tokens": 17484747.0,
      "step": 7980
    },
    {
      "entropy": 5.541376209259033,
      "epoch": 0.7924771734815403,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000494578952448766,
      "loss": 5.2792,
      "mean_token_accuracy": 0.15907120555639268,
      "num_tokens": 17496299.0,
      "step": 7985
    },
    {
      "entropy": 5.604167509078979,
      "epoch": 0.7929734021437078,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004945712189513161,
      "loss": 5.441,
      "mean_token_accuracy": 0.14787812530994415,
      "num_tokens": 17507448.0,
      "step": 7990
    },
    {
      "entropy": 5.628121995925904,
      "epoch": 0.7934696308058753,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004945634800089733,
      "loss": 5.4042,
      "mean_token_accuracy": 0.15441112518310546,
      "num_tokens": 17520112.0,
      "step": 7995
    },
    {
      "entropy": 5.613991737365723,
      "epoch": 0.7939658594680429,
      "grad_norm": 0.875,
      "learning_rate": 0.0004945557356219291,
      "loss": 5.4044,
      "mean_token_accuracy": 0.14858510717749596,
      "num_tokens": 17531368.0,
      "step": 8000
    },
    {
      "entropy": 5.536077356338501,
      "epoch": 0.7944620881302104,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004945479857903758,
      "loss": 5.3431,
      "mean_token_accuracy": 0.15966367572546006,
      "num_tokens": 17542082.0,
      "step": 8005
    },
    {
      "entropy": 5.62625732421875,
      "epoch": 0.7949583167923779,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004945402305145056,
      "loss": 5.38,
      "mean_token_accuracy": 0.15365598797798158,
      "num_tokens": 17555518.0,
      "step": 8010
    },
    {
      "entropy": 5.577501440048218,
      "epoch": 0.7954545454545454,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004945324697945107,
      "loss": 5.3423,
      "mean_token_accuracy": 0.1595192164182663,
      "num_tokens": 17566110.0,
      "step": 8015
    },
    {
      "entropy": 5.473816680908203,
      "epoch": 0.795950774116713,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004945247036305836,
      "loss": 5.2663,
      "mean_token_accuracy": 0.1575251117348671,
      "num_tokens": 17575942.0,
      "step": 8020
    },
    {
      "entropy": 5.535625362396241,
      "epoch": 0.7964470027788805,
      "grad_norm": 1.0,
      "learning_rate": 0.0004945169320229169,
      "loss": 5.3364,
      "mean_token_accuracy": 0.162310454249382,
      "num_tokens": 17586001.0,
      "step": 8025
    },
    {
      "entropy": 5.641378355026245,
      "epoch": 0.7969432314410481,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004945091549717033,
      "loss": 5.4955,
      "mean_token_accuracy": 0.14408404529094695,
      "num_tokens": 17596774.0,
      "step": 8030
    },
    {
      "entropy": 5.5768969535827635,
      "epoch": 0.7974394601032155,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004945013724771356,
      "loss": 5.344,
      "mean_token_accuracy": 0.15899086594581605,
      "num_tokens": 17606557.0,
      "step": 8035
    },
    {
      "entropy": 5.46403317451477,
      "epoch": 0.7979356887653831,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004944935845394069,
      "loss": 5.2091,
      "mean_token_accuracy": 0.16562122851610184,
      "num_tokens": 17616465.0,
      "step": 8040
    },
    {
      "entropy": 5.579554080963135,
      "epoch": 0.7984319174275506,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004944857911587103,
      "loss": 5.3864,
      "mean_token_accuracy": 0.15155401676893235,
      "num_tokens": 17626907.0,
      "step": 8045
    },
    {
      "entropy": 5.6786962985992435,
      "epoch": 0.7989281460897182,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004944779923352389,
      "loss": 5.3975,
      "mean_token_accuracy": 0.15766018629074097,
      "num_tokens": 17638660.0,
      "step": 8050
    },
    {
      "entropy": 5.517003345489502,
      "epoch": 0.7994243747518857,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004944701880691864,
      "loss": 5.3138,
      "mean_token_accuracy": 0.15925874710083007,
      "num_tokens": 17650135.0,
      "step": 8055
    },
    {
      "entropy": 5.460254287719726,
      "epoch": 0.7999206034140532,
      "grad_norm": 0.90625,
      "learning_rate": 0.000494462378360746,
      "loss": 5.3311,
      "mean_token_accuracy": 0.1636511966586113,
      "num_tokens": 17660987.0,
      "step": 8060
    },
    {
      "entropy": 5.543812704086304,
      "epoch": 0.8004168320762207,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004944545632101117,
      "loss": 5.2574,
      "mean_token_accuracy": 0.15987827330827714,
      "num_tokens": 17672520.0,
      "step": 8065
    },
    {
      "entropy": 5.575038909912109,
      "epoch": 0.8009130607383882,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004944467426174769,
      "loss": 5.3339,
      "mean_token_accuracy": 0.1598665416240692,
      "num_tokens": 17683688.0,
      "step": 8070
    },
    {
      "entropy": 5.50664210319519,
      "epoch": 0.8014092894005558,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000494438916583036,
      "loss": 5.2699,
      "mean_token_accuracy": 0.16346991956233978,
      "num_tokens": 17695213.0,
      "step": 8075
    },
    {
      "entropy": 5.575667428970337,
      "epoch": 0.8019055180627233,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004944310851069826,
      "loss": 5.3486,
      "mean_token_accuracy": 0.15848584920167924,
      "num_tokens": 17705545.0,
      "step": 8080
    },
    {
      "entropy": 5.558717441558838,
      "epoch": 0.8024017467248908,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004944232481895112,
      "loss": 5.3209,
      "mean_token_accuracy": 0.1664901480078697,
      "num_tokens": 17715855.0,
      "step": 8085
    },
    {
      "entropy": 5.614819765090942,
      "epoch": 0.8028979753870583,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000494415405830816,
      "loss": 5.4158,
      "mean_token_accuracy": 0.15779888331890107,
      "num_tokens": 17727764.0,
      "step": 8090
    },
    {
      "entropy": 5.56912522315979,
      "epoch": 0.8033942040492259,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004944075580310916,
      "loss": 5.398,
      "mean_token_accuracy": 0.15286846309900284,
      "num_tokens": 17738464.0,
      "step": 8095
    },
    {
      "entropy": 5.577541494369507,
      "epoch": 0.8038904327113934,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004943997047905325,
      "loss": 5.3703,
      "mean_token_accuracy": 0.1655590355396271,
      "num_tokens": 17747502.0,
      "step": 8100
    },
    {
      "entropy": 5.551443243026734,
      "epoch": 0.804386661373561,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004943918461093335,
      "loss": 5.367,
      "mean_token_accuracy": 0.1538964331150055,
      "num_tokens": 17759520.0,
      "step": 8105
    },
    {
      "entropy": 5.5665810108184814,
      "epoch": 0.8048828900357284,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004943839819876894,
      "loss": 5.3329,
      "mean_token_accuracy": 0.1593322053551674,
      "num_tokens": 17770631.0,
      "step": 8110
    },
    {
      "entropy": 5.602841758728028,
      "epoch": 0.805379118697896,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004943761124257954,
      "loss": 5.2946,
      "mean_token_accuracy": 0.16228071600198746,
      "num_tokens": 17781320.0,
      "step": 8115
    },
    {
      "entropy": 5.590414524078369,
      "epoch": 0.8058753473600635,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004943682374238464,
      "loss": 5.3649,
      "mean_token_accuracy": 0.15804662331938743,
      "num_tokens": 17792627.0,
      "step": 8120
    },
    {
      "entropy": 5.540787553787231,
      "epoch": 0.8063715760222311,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000494360356982038,
      "loss": 5.406,
      "mean_token_accuracy": 0.15668713301420212,
      "num_tokens": 17803286.0,
      "step": 8125
    },
    {
      "entropy": 5.480075263977051,
      "epoch": 0.8068678046843986,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004943524711005653,
      "loss": 5.2015,
      "mean_token_accuracy": 0.16303706616163255,
      "num_tokens": 17813631.0,
      "step": 8130
    },
    {
      "entropy": 5.504738998413086,
      "epoch": 0.8073640333465661,
      "grad_norm": 0.953125,
      "learning_rate": 0.000494344579779624,
      "loss": 5.3063,
      "mean_token_accuracy": 0.16077257096767425,
      "num_tokens": 17823467.0,
      "step": 8135
    },
    {
      "entropy": 5.5465771675109865,
      "epoch": 0.8078602620087336,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004943366830194098,
      "loss": 5.307,
      "mean_token_accuracy": 0.16680953353643418,
      "num_tokens": 17833600.0,
      "step": 8140
    },
    {
      "entropy": 5.558188152313233,
      "epoch": 0.8083564906709012,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004943287808201183,
      "loss": 5.2986,
      "mean_token_accuracy": 0.15690502524375916,
      "num_tokens": 17843112.0,
      "step": 8145
    },
    {
      "entropy": 5.561684846878052,
      "epoch": 0.8088527193330687,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004943208731819459,
      "loss": 5.3072,
      "mean_token_accuracy": 0.15987997353076935,
      "num_tokens": 17853678.0,
      "step": 8150
    },
    {
      "entropy": 5.5255622386932375,
      "epoch": 0.8093489479952362,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004943129601050883,
      "loss": 5.2885,
      "mean_token_accuracy": 0.1580919548869133,
      "num_tokens": 17863537.0,
      "step": 8155
    },
    {
      "entropy": 5.550281381607055,
      "epoch": 0.8098451766574037,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000494305041589742,
      "loss": 5.3723,
      "mean_token_accuracy": 0.15576187670230865,
      "num_tokens": 17873511.0,
      "step": 8160
    },
    {
      "entropy": 5.533975887298584,
      "epoch": 0.8103414053195712,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004942971176361031,
      "loss": 5.2837,
      "mean_token_accuracy": 0.16323758363723756,
      "num_tokens": 17884546.0,
      "step": 8165
    },
    {
      "entropy": 5.526020526885986,
      "epoch": 0.8108376339817388,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004942891882443682,
      "loss": 5.2443,
      "mean_token_accuracy": 0.16286058723926544,
      "num_tokens": 17895331.0,
      "step": 8170
    },
    {
      "entropy": 5.571474981307984,
      "epoch": 0.8113338626439063,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000494281253414734,
      "loss": 5.4073,
      "mean_token_accuracy": 0.15418355911970139,
      "num_tokens": 17906465.0,
      "step": 8175
    },
    {
      "entropy": 5.532446432113647,
      "epoch": 0.8118300913060739,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004942733131473973,
      "loss": 5.3441,
      "mean_token_accuracy": 0.15361143350601197,
      "num_tokens": 17916936.0,
      "step": 8180
    },
    {
      "entropy": 5.652284479141235,
      "epoch": 0.8123263199682413,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004942653674425548,
      "loss": 5.3573,
      "mean_token_accuracy": 0.1522192656993866,
      "num_tokens": 17927522.0,
      "step": 8185
    },
    {
      "entropy": 5.616598796844483,
      "epoch": 0.8128225486304089,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004942574163004038,
      "loss": 5.4177,
      "mean_token_accuracy": 0.15336811244487764,
      "num_tokens": 17938216.0,
      "step": 8190
    },
    {
      "entropy": 5.556334447860718,
      "epoch": 0.8133187772925764,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004942494597211412,
      "loss": 5.3068,
      "mean_token_accuracy": 0.15798144042491913,
      "num_tokens": 17949333.0,
      "step": 8195
    },
    {
      "entropy": 5.6230260848999025,
      "epoch": 0.813815005954744,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004942414977049646,
      "loss": 5.3813,
      "mean_token_accuracy": 0.14451824128627777,
      "num_tokens": 17960178.0,
      "step": 8200
    },
    {
      "entropy": 5.552642488479615,
      "epoch": 0.8143112346169115,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000494233530252071,
      "loss": 5.3371,
      "mean_token_accuracy": 0.15979109555482865,
      "num_tokens": 17971180.0,
      "step": 8205
    },
    {
      "entropy": 5.567732620239258,
      "epoch": 0.814807463279079,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004942255573626585,
      "loss": 5.3655,
      "mean_token_accuracy": 0.1580875739455223,
      "num_tokens": 17982572.0,
      "step": 8210
    },
    {
      "entropy": 5.600094747543335,
      "epoch": 0.8153036919412465,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004942175790369244,
      "loss": 5.2915,
      "mean_token_accuracy": 0.16242016702890397,
      "num_tokens": 17993236.0,
      "step": 8215
    },
    {
      "entropy": 5.568907976150513,
      "epoch": 0.8157999206034141,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004942095952750669,
      "loss": 5.4123,
      "mean_token_accuracy": 0.15650524497032164,
      "num_tokens": 18004215.0,
      "step": 8220
    },
    {
      "entropy": 5.543283700942993,
      "epoch": 0.8162961492655816,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004942016060772837,
      "loss": 5.3112,
      "mean_token_accuracy": 0.16373128592967987,
      "num_tokens": 18013880.0,
      "step": 8225
    },
    {
      "entropy": 5.578078413009644,
      "epoch": 0.8167923779277491,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004941936114437731,
      "loss": 5.3558,
      "mean_token_accuracy": 0.1561330273747444,
      "num_tokens": 18024928.0,
      "step": 8230
    },
    {
      "entropy": 5.560678339004516,
      "epoch": 0.8172886065899166,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004941856113747331,
      "loss": 5.3184,
      "mean_token_accuracy": 0.15780884623527527,
      "num_tokens": 18035857.0,
      "step": 8235
    },
    {
      "entropy": 5.629302167892456,
      "epoch": 0.8177848352520841,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004941776058703623,
      "loss": 5.4277,
      "mean_token_accuracy": 0.15591926872730255,
      "num_tokens": 18046776.0,
      "step": 8240
    },
    {
      "entropy": 5.582234573364258,
      "epoch": 0.8182810639142517,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004941695949308593,
      "loss": 5.3828,
      "mean_token_accuracy": 0.16022476106882094,
      "num_tokens": 18057167.0,
      "step": 8245
    },
    {
      "entropy": 5.578600931167602,
      "epoch": 0.8187772925764192,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004941615785564224,
      "loss": 5.3976,
      "mean_token_accuracy": 0.15019726008176804,
      "num_tokens": 18066953.0,
      "step": 8250
    },
    {
      "entropy": 5.55739197731018,
      "epoch": 0.8192735212385868,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004941535567472506,
      "loss": 5.2314,
      "mean_token_accuracy": 0.16082544922828673,
      "num_tokens": 18077977.0,
      "step": 8255
    },
    {
      "entropy": 5.53443341255188,
      "epoch": 0.8197697499007542,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000494145529503543,
      "loss": 5.3292,
      "mean_token_accuracy": 0.15666717290878296,
      "num_tokens": 18087581.0,
      "step": 8260
    },
    {
      "entropy": 5.475851774215698,
      "epoch": 0.8202659785629218,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004941374968254985,
      "loss": 5.2356,
      "mean_token_accuracy": 0.1597230166196823,
      "num_tokens": 18098193.0,
      "step": 8265
    },
    {
      "entropy": 5.548586225509643,
      "epoch": 0.8207622072250893,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004941294587133161,
      "loss": 5.3556,
      "mean_token_accuracy": 0.15848664790391923,
      "num_tokens": 18110830.0,
      "step": 8270
    },
    {
      "entropy": 5.597237873077392,
      "epoch": 0.8212584358872569,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004941214151671954,
      "loss": 5.3835,
      "mean_token_accuracy": 0.15455053448677064,
      "num_tokens": 18122073.0,
      "step": 8275
    },
    {
      "entropy": 5.521269798278809,
      "epoch": 0.8217546645494244,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004941133661873357,
      "loss": 5.3386,
      "mean_token_accuracy": 0.15720982998609542,
      "num_tokens": 18132777.0,
      "step": 8280
    },
    {
      "entropy": 5.478314447402954,
      "epoch": 0.822250893211592,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004941053117739367,
      "loss": 5.2657,
      "mean_token_accuracy": 0.15625037103891373,
      "num_tokens": 18143232.0,
      "step": 8285
    },
    {
      "entropy": 5.552276754379273,
      "epoch": 0.8227471218737594,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004940972519271982,
      "loss": 5.4458,
      "mean_token_accuracy": 0.15443039685487747,
      "num_tokens": 18155225.0,
      "step": 8290
    },
    {
      "entropy": 5.566086196899414,
      "epoch": 0.823243350535927,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004940891866473199,
      "loss": 5.3689,
      "mean_token_accuracy": 0.15550159364938737,
      "num_tokens": 18165945.0,
      "step": 8295
    },
    {
      "entropy": 5.621346950531006,
      "epoch": 0.8237395791980945,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004940811159345017,
      "loss": 5.4121,
      "mean_token_accuracy": 0.1536681219935417,
      "num_tokens": 18177271.0,
      "step": 8300
    },
    {
      "entropy": 5.600881147384643,
      "epoch": 0.824235807860262,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004940730397889441,
      "loss": 5.3436,
      "mean_token_accuracy": 0.15792988687753678,
      "num_tokens": 18189732.0,
      "step": 8305
    },
    {
      "entropy": 5.562526035308838,
      "epoch": 0.8247320365224295,
      "grad_norm": 0.875,
      "learning_rate": 0.0004940649582108472,
      "loss": 5.4431,
      "mean_token_accuracy": 0.15375415980815887,
      "num_tokens": 18200731.0,
      "step": 8310
    },
    {
      "entropy": 5.58730411529541,
      "epoch": 0.825228265184597,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004940568712004113,
      "loss": 5.3553,
      "mean_token_accuracy": 0.15794347524642943,
      "num_tokens": 18210873.0,
      "step": 8315
    },
    {
      "entropy": 5.621802616119385,
      "epoch": 0.8257244938467646,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004940487787578369,
      "loss": 5.3582,
      "mean_token_accuracy": 0.15885585248470308,
      "num_tokens": 18222072.0,
      "step": 8320
    },
    {
      "entropy": 5.510449028015136,
      "epoch": 0.8262207225089321,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000494040680883325,
      "loss": 5.2685,
      "mean_token_accuracy": 0.15838711857795715,
      "num_tokens": 18232522.0,
      "step": 8325
    },
    {
      "entropy": 5.585622549057007,
      "epoch": 0.8267169511710997,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004940325775770762,
      "loss": 5.3562,
      "mean_token_accuracy": 0.152874194085598,
      "num_tokens": 18243587.0,
      "step": 8330
    },
    {
      "entropy": 5.499403047561645,
      "epoch": 0.8272131798332671,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004940244688392913,
      "loss": 5.2735,
      "mean_token_accuracy": 0.16457149386405945,
      "num_tokens": 18254965.0,
      "step": 8335
    },
    {
      "entropy": 5.526314306259155,
      "epoch": 0.8277094084954347,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004940163546701716,
      "loss": 5.304,
      "mean_token_accuracy": 0.1621924877166748,
      "num_tokens": 18265892.0,
      "step": 8340
    },
    {
      "entropy": 5.533067941665649,
      "epoch": 0.8282056371576022,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004940082350699183,
      "loss": 5.3054,
      "mean_token_accuracy": 0.1597759246826172,
      "num_tokens": 18276713.0,
      "step": 8345
    },
    {
      "entropy": 5.6015809059143065,
      "epoch": 0.8287018658197698,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004940001100387325,
      "loss": 5.3487,
      "mean_token_accuracy": 0.15081969648599625,
      "num_tokens": 18288846.0,
      "step": 8350
    },
    {
      "entropy": 5.594089126586914,
      "epoch": 0.8291980944819373,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004939919795768161,
      "loss": 5.3696,
      "mean_token_accuracy": 0.16012121886014938,
      "num_tokens": 18298750.0,
      "step": 8355
    },
    {
      "entropy": 5.524968576431275,
      "epoch": 0.8296943231441049,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939838436843704,
      "loss": 5.3483,
      "mean_token_accuracy": 0.15921227037906646,
      "num_tokens": 18310065.0,
      "step": 8360
    },
    {
      "entropy": 5.545454597473144,
      "epoch": 0.8301905518062723,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004939757023615973,
      "loss": 5.2898,
      "mean_token_accuracy": 0.15805892199277877,
      "num_tokens": 18319498.0,
      "step": 8365
    },
    {
      "entropy": 5.4899983406066895,
      "epoch": 0.8306867804684399,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004939675556086986,
      "loss": 5.2651,
      "mean_token_accuracy": 0.16524300426244737,
      "num_tokens": 18329718.0,
      "step": 8370
    },
    {
      "entropy": 5.522244691848755,
      "epoch": 0.8311830091306074,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939594034258763,
      "loss": 5.2852,
      "mean_token_accuracy": 0.16079603731632233,
      "num_tokens": 18341628.0,
      "step": 8375
    },
    {
      "entropy": 5.5233237743377686,
      "epoch": 0.8316792377927749,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004939512458133326,
      "loss": 5.3232,
      "mean_token_accuracy": 0.1642431765794754,
      "num_tokens": 18352869.0,
      "step": 8380
    },
    {
      "entropy": 5.534032392501831,
      "epoch": 0.8321754664549424,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004939430827712698,
      "loss": 5.3238,
      "mean_token_accuracy": 0.15860947966575623,
      "num_tokens": 18363664.0,
      "step": 8385
    },
    {
      "entropy": 5.594780349731446,
      "epoch": 0.8326716951171099,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004939349142998905,
      "loss": 5.3113,
      "mean_token_accuracy": 0.15810291171073915,
      "num_tokens": 18374325.0,
      "step": 8390
    },
    {
      "entropy": 5.5217423915863035,
      "epoch": 0.8331679237792775,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004939267403993968,
      "loss": 5.2966,
      "mean_token_accuracy": 0.1612369954586029,
      "num_tokens": 18384547.0,
      "step": 8395
    },
    {
      "entropy": 5.551669120788574,
      "epoch": 0.833664152441445,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000493918561069992,
      "loss": 5.2874,
      "mean_token_accuracy": 0.16169873923063277,
      "num_tokens": 18395141.0,
      "step": 8400
    },
    {
      "entropy": 5.522643756866455,
      "epoch": 0.8341603811036126,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004939103763118784,
      "loss": 5.2917,
      "mean_token_accuracy": 0.15762884318828582,
      "num_tokens": 18406326.0,
      "step": 8405
    },
    {
      "entropy": 5.537695217132568,
      "epoch": 0.83465660976578,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004939021861252593,
      "loss": 5.3415,
      "mean_token_accuracy": 0.16229671239852905,
      "num_tokens": 18417521.0,
      "step": 8410
    },
    {
      "entropy": 5.561509704589843,
      "epoch": 0.8351528384279476,
      "grad_norm": 1.0,
      "learning_rate": 0.0004938939905103376,
      "loss": 5.277,
      "mean_token_accuracy": 0.15979011803865434,
      "num_tokens": 18426813.0,
      "step": 8415
    },
    {
      "entropy": 5.5146078109741214,
      "epoch": 0.8356490670901151,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004938857894673166,
      "loss": 5.2444,
      "mean_token_accuracy": 0.1613193228840828,
      "num_tokens": 18438227.0,
      "step": 8420
    },
    {
      "entropy": 5.525485038757324,
      "epoch": 0.8361452957522827,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004938775829963997,
      "loss": 5.3727,
      "mean_token_accuracy": 0.15905213952064515,
      "num_tokens": 18448915.0,
      "step": 8425
    },
    {
      "entropy": 5.57030463218689,
      "epoch": 0.8366415244144502,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004938693710977904,
      "loss": 5.2672,
      "mean_token_accuracy": 0.16031362116336823,
      "num_tokens": 18459330.0,
      "step": 8430
    },
    {
      "entropy": 5.532431697845459,
      "epoch": 0.8371377530766178,
      "grad_norm": 0.875,
      "learning_rate": 0.0004938611537716922,
      "loss": 5.2386,
      "mean_token_accuracy": 0.16233667135238647,
      "num_tokens": 18469813.0,
      "step": 8435
    },
    {
      "entropy": 5.490002584457398,
      "epoch": 0.8376339817387852,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004938529310183091,
      "loss": 5.2947,
      "mean_token_accuracy": 0.156914222240448,
      "num_tokens": 18479783.0,
      "step": 8440
    },
    {
      "entropy": 5.510782670974732,
      "epoch": 0.8381302104009528,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000493844702837845,
      "loss": 5.3036,
      "mean_token_accuracy": 0.15965810865163804,
      "num_tokens": 18489640.0,
      "step": 8445
    },
    {
      "entropy": 5.560747814178467,
      "epoch": 0.8386264390631203,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004938364692305036,
      "loss": 5.3664,
      "mean_token_accuracy": 0.1613190710544586,
      "num_tokens": 18500744.0,
      "step": 8450
    },
    {
      "entropy": 5.55366849899292,
      "epoch": 0.8391226677252878,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004938282301964896,
      "loss": 5.2796,
      "mean_token_accuracy": 0.16003650277853013,
      "num_tokens": 18513286.0,
      "step": 8455
    },
    {
      "entropy": 5.599881029129028,
      "epoch": 0.8396188963874553,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004938199857360068,
      "loss": 5.2563,
      "mean_token_accuracy": 0.16317568719387054,
      "num_tokens": 18522934.0,
      "step": 8460
    },
    {
      "entropy": 5.523234224319458,
      "epoch": 0.8401151250496228,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00049381173584926,
      "loss": 5.3165,
      "mean_token_accuracy": 0.16004646569490433,
      "num_tokens": 18534828.0,
      "step": 8465
    },
    {
      "entropy": 5.494680166244507,
      "epoch": 0.8406113537117904,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004938034805364535,
      "loss": 5.2605,
      "mean_token_accuracy": 0.16069284826517105,
      "num_tokens": 18544733.0,
      "step": 8470
    },
    {
      "entropy": 5.531710147857666,
      "epoch": 0.8411075823739579,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004937952197977923,
      "loss": 5.3305,
      "mean_token_accuracy": 0.15478749722242355,
      "num_tokens": 18554583.0,
      "step": 8475
    },
    {
      "entropy": 5.6344075202941895,
      "epoch": 0.8416038110361255,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000493786953633481,
      "loss": 5.3668,
      "mean_token_accuracy": 0.15707846879959106,
      "num_tokens": 18566503.0,
      "step": 8480
    },
    {
      "entropy": 5.5227796077728275,
      "epoch": 0.842100039698293,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004937786820437248,
      "loss": 5.3151,
      "mean_token_accuracy": 0.1600247323513031,
      "num_tokens": 18578161.0,
      "step": 8485
    },
    {
      "entropy": 5.525348567962647,
      "epoch": 0.8425962683604605,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004937704050287286,
      "loss": 5.3631,
      "mean_token_accuracy": 0.16227232664823532,
      "num_tokens": 18588212.0,
      "step": 8490
    },
    {
      "entropy": 5.668597173690796,
      "epoch": 0.843092497022628,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004937621225886978,
      "loss": 5.3781,
      "mean_token_accuracy": 0.15517351925373077,
      "num_tokens": 18601293.0,
      "step": 8495
    },
    {
      "entropy": 5.592793798446655,
      "epoch": 0.8435887256847956,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004937538347238378,
      "loss": 5.4242,
      "mean_token_accuracy": 0.15348978191614152,
      "num_tokens": 18613316.0,
      "step": 8500
    },
    {
      "entropy": 5.537801313400268,
      "epoch": 0.8440849543469631,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000493745541434354,
      "loss": 5.3559,
      "mean_token_accuracy": 0.1587686210870743,
      "num_tokens": 18624461.0,
      "step": 8505
    },
    {
      "entropy": 5.5556048393249515,
      "epoch": 0.8445811830091307,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004937372427204522,
      "loss": 5.2578,
      "mean_token_accuracy": 0.16012921929359436,
      "num_tokens": 18635165.0,
      "step": 8510
    },
    {
      "entropy": 5.566375350952148,
      "epoch": 0.8450774116712981,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004937289385823382,
      "loss": 5.3067,
      "mean_token_accuracy": 0.16270599886775017,
      "num_tokens": 18646649.0,
      "step": 8515
    },
    {
      "entropy": 5.5332996368408205,
      "epoch": 0.8455736403334657,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004937206290202176,
      "loss": 5.2953,
      "mean_token_accuracy": 0.15720799714326858,
      "num_tokens": 18658527.0,
      "step": 8520
    },
    {
      "entropy": 5.535870552062988,
      "epoch": 0.8460698689956332,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004937123140342969,
      "loss": 5.2992,
      "mean_token_accuracy": 0.15728411376476287,
      "num_tokens": 18669463.0,
      "step": 8525
    },
    {
      "entropy": 5.582309579849243,
      "epoch": 0.8465660976578007,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000493703993624782,
      "loss": 5.3422,
      "mean_token_accuracy": 0.1569891206920147,
      "num_tokens": 18680806.0,
      "step": 8530
    },
    {
      "entropy": 5.553014659881592,
      "epoch": 0.8470623263199683,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004936956677918794,
      "loss": 5.325,
      "mean_token_accuracy": 0.15743035674095154,
      "num_tokens": 18692124.0,
      "step": 8535
    },
    {
      "entropy": 5.593314981460571,
      "epoch": 0.8475585549821357,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004936873365357955,
      "loss": 5.3591,
      "mean_token_accuracy": 0.15825066715478897,
      "num_tokens": 18702017.0,
      "step": 8540
    },
    {
      "entropy": 5.466948652267456,
      "epoch": 0.8480547836443033,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004936789998567368,
      "loss": 5.2504,
      "mean_token_accuracy": 0.1651918262243271,
      "num_tokens": 18713205.0,
      "step": 8545
    },
    {
      "entropy": 5.537358236312866,
      "epoch": 0.8485510123064708,
      "grad_norm": 0.875,
      "learning_rate": 0.0004936706577549101,
      "loss": 5.2623,
      "mean_token_accuracy": 0.1572915345430374,
      "num_tokens": 18724165.0,
      "step": 8550
    },
    {
      "entropy": 5.533035326004028,
      "epoch": 0.8490472409686384,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004936623102305223,
      "loss": 5.3234,
      "mean_token_accuracy": 0.1630731612443924,
      "num_tokens": 18735155.0,
      "step": 8555
    },
    {
      "entropy": 5.5569404602050785,
      "epoch": 0.8495434696308058,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004936539572837804,
      "loss": 5.4208,
      "mean_token_accuracy": 0.15740315914154052,
      "num_tokens": 18746228.0,
      "step": 8560
    },
    {
      "entropy": 5.556674766540527,
      "epoch": 0.8500396982929734,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004936455989148916,
      "loss": 5.3713,
      "mean_token_accuracy": 0.15661710053682326,
      "num_tokens": 18756638.0,
      "step": 8565
    },
    {
      "entropy": 5.604758405685425,
      "epoch": 0.8505359269551409,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004936372351240629,
      "loss": 5.3341,
      "mean_token_accuracy": 0.16325204819440842,
      "num_tokens": 18768651.0,
      "step": 8570
    },
    {
      "entropy": 5.6614268779754635,
      "epoch": 0.8510321556173085,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000493628865911502,
      "loss": 5.4427,
      "mean_token_accuracy": 0.15434404462575912,
      "num_tokens": 18780399.0,
      "step": 8575
    },
    {
      "entropy": 5.519382762908935,
      "epoch": 0.851528384279476,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004936204912774163,
      "loss": 5.3044,
      "mean_token_accuracy": 0.1619669407606125,
      "num_tokens": 18791927.0,
      "step": 8580
    },
    {
      "entropy": 5.490979290008545,
      "epoch": 0.8520246129416436,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004936121112220135,
      "loss": 5.2797,
      "mean_token_accuracy": 0.15910061895847322,
      "num_tokens": 18802779.0,
      "step": 8585
    },
    {
      "entropy": 5.4997974872589115,
      "epoch": 0.852520841603811,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004936037257455015,
      "loss": 5.2769,
      "mean_token_accuracy": 0.15892841964960097,
      "num_tokens": 18813123.0,
      "step": 8590
    },
    {
      "entropy": 5.527001857757568,
      "epoch": 0.8530170702659786,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000493595334848088,
      "loss": 5.2799,
      "mean_token_accuracy": 0.16557230204343795,
      "num_tokens": 18823699.0,
      "step": 8595
    },
    {
      "entropy": 5.614219713211059,
      "epoch": 0.8535132989281461,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004935869385299813,
      "loss": 5.4374,
      "mean_token_accuracy": 0.15188810378313064,
      "num_tokens": 18835450.0,
      "step": 8600
    },
    {
      "entropy": 5.550320196151733,
      "epoch": 0.8540095275903136,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004935785367913896,
      "loss": 5.2802,
      "mean_token_accuracy": 0.15528021305799483,
      "num_tokens": 18845919.0,
      "step": 8605
    },
    {
      "entropy": 5.5714820384979244,
      "epoch": 0.8545057562524812,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004935701296325212,
      "loss": 5.4018,
      "mean_token_accuracy": 0.15196267366409302,
      "num_tokens": 18858376.0,
      "step": 8610
    },
    {
      "entropy": 5.567236423492432,
      "epoch": 0.8550019849146486,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004935617170535845,
      "loss": 5.357,
      "mean_token_accuracy": 0.16111219227313994,
      "num_tokens": 18870456.0,
      "step": 8615
    },
    {
      "entropy": 5.539654159545899,
      "epoch": 0.8554982135768162,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004935532990547882,
      "loss": 5.3669,
      "mean_token_accuracy": 0.15522739589214324,
      "num_tokens": 18882174.0,
      "step": 8620
    },
    {
      "entropy": 5.640656852722168,
      "epoch": 0.8559944422389837,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004935448756363409,
      "loss": 5.4789,
      "mean_token_accuracy": 0.15537660121917723,
      "num_tokens": 18892982.0,
      "step": 8625
    },
    {
      "entropy": 5.610037469863892,
      "epoch": 0.8564906709011513,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004935364467984518,
      "loss": 5.2944,
      "mean_token_accuracy": 0.16325953602790833,
      "num_tokens": 18903557.0,
      "step": 8630
    },
    {
      "entropy": 5.615718507766724,
      "epoch": 0.8569868995633187,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004935280125413295,
      "loss": 5.4494,
      "mean_token_accuracy": 0.15439070388674736,
      "num_tokens": 18914912.0,
      "step": 8635
    },
    {
      "entropy": 5.599884176254273,
      "epoch": 0.8574831282254863,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004935195728651836,
      "loss": 5.3884,
      "mean_token_accuracy": 0.15769529342651367,
      "num_tokens": 18927773.0,
      "step": 8640
    },
    {
      "entropy": 5.562195873260498,
      "epoch": 0.8579793568876538,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000493511127770223,
      "loss": 5.3292,
      "mean_token_accuracy": 0.15460922122001647,
      "num_tokens": 18938313.0,
      "step": 8645
    },
    {
      "entropy": 5.537086153030396,
      "epoch": 0.8584755855498214,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004935026772566574,
      "loss": 5.2694,
      "mean_token_accuracy": 0.16621735841035842,
      "num_tokens": 18948600.0,
      "step": 8650
    },
    {
      "entropy": 5.6071208953857425,
      "epoch": 0.8589718142119889,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004934942213246961,
      "loss": 5.366,
      "mean_token_accuracy": 0.16027262061834335,
      "num_tokens": 18958927.0,
      "step": 8655
    },
    {
      "entropy": 5.541692781448364,
      "epoch": 0.8594680428741565,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000493485759974549,
      "loss": 5.3137,
      "mean_token_accuracy": 0.15310209542512893,
      "num_tokens": 18968625.0,
      "step": 8660
    },
    {
      "entropy": 5.471400356292724,
      "epoch": 0.8599642715363239,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004934772932064259,
      "loss": 5.2326,
      "mean_token_accuracy": 0.1595799967646599,
      "num_tokens": 18979772.0,
      "step": 8665
    },
    {
      "entropy": 5.564545679092407,
      "epoch": 0.8604605001984915,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004934688210205367,
      "loss": 5.3956,
      "mean_token_accuracy": 0.15682338774204255,
      "num_tokens": 18991522.0,
      "step": 8670
    },
    {
      "entropy": 5.575645589828492,
      "epoch": 0.860956728860659,
      "grad_norm": 0.875,
      "learning_rate": 0.0004934603434170913,
      "loss": 5.3541,
      "mean_token_accuracy": 0.1573867157101631,
      "num_tokens": 19002479.0,
      "step": 8675
    },
    {
      "entropy": 5.603612327575684,
      "epoch": 0.8614529575228265,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004934518603963003,
      "loss": 5.4179,
      "mean_token_accuracy": 0.15240964889526368,
      "num_tokens": 19013254.0,
      "step": 8680
    },
    {
      "entropy": 5.524136972427368,
      "epoch": 0.861949186184994,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004934433719583739,
      "loss": 5.2423,
      "mean_token_accuracy": 0.16270106732845308,
      "num_tokens": 19023943.0,
      "step": 8685
    },
    {
      "entropy": 5.619406223297119,
      "epoch": 0.8624454148471615,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004934348781035226,
      "loss": 5.3791,
      "mean_token_accuracy": 0.15710915327072145,
      "num_tokens": 19034672.0,
      "step": 8690
    },
    {
      "entropy": 5.626100206375122,
      "epoch": 0.8629416435093291,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493426378831957,
      "loss": 5.3601,
      "mean_token_accuracy": 0.16364834159612657,
      "num_tokens": 19045758.0,
      "step": 8695
    },
    {
      "entropy": 5.546971464157105,
      "epoch": 0.8634378721714966,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004934178741438877,
      "loss": 5.3529,
      "mean_token_accuracy": 0.16413892656564713,
      "num_tokens": 19057526.0,
      "step": 8700
    },
    {
      "entropy": 5.49228720664978,
      "epoch": 0.8639341008336642,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000493409364039526,
      "loss": 5.2166,
      "mean_token_accuracy": 0.16375850439071654,
      "num_tokens": 19069657.0,
      "step": 8705
    },
    {
      "entropy": 5.535357713699341,
      "epoch": 0.8644303294958317,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004934008485190825,
      "loss": 5.2839,
      "mean_token_accuracy": 0.16336794346570968,
      "num_tokens": 19080699.0,
      "step": 8710
    },
    {
      "entropy": 5.536669921875,
      "epoch": 0.8649265581579992,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004933923275827687,
      "loss": 5.3222,
      "mean_token_accuracy": 0.15725933462381364,
      "num_tokens": 19093041.0,
      "step": 8715
    },
    {
      "entropy": 5.627826738357544,
      "epoch": 0.8654227868201667,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004933838012307957,
      "loss": 5.3765,
      "mean_token_accuracy": 0.1537148669362068,
      "num_tokens": 19104249.0,
      "step": 8720
    },
    {
      "entropy": 5.579020833969116,
      "epoch": 0.8659190154823343,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000493375269463375,
      "loss": 5.3417,
      "mean_token_accuracy": 0.1575908273458481,
      "num_tokens": 19114434.0,
      "step": 8725
    },
    {
      "entropy": 5.537295150756836,
      "epoch": 0.8664152441445018,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004933667322807182,
      "loss": 5.3558,
      "mean_token_accuracy": 0.15704239457845687,
      "num_tokens": 19124814.0,
      "step": 8730
    },
    {
      "entropy": 5.565311050415039,
      "epoch": 0.8669114728066694,
      "grad_norm": 0.9375,
      "learning_rate": 0.000493358189683037,
      "loss": 5.372,
      "mean_token_accuracy": 0.15405240654945374,
      "num_tokens": 19135205.0,
      "step": 8735
    },
    {
      "entropy": 5.5298460006713865,
      "epoch": 0.8674077014688368,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004933496416705432,
      "loss": 5.2593,
      "mean_token_accuracy": 0.1580391377210617,
      "num_tokens": 19147117.0,
      "step": 8740
    },
    {
      "entropy": 5.5313962459564205,
      "epoch": 0.8679039301310044,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004933410882434488,
      "loss": 5.3397,
      "mean_token_accuracy": 0.162195847928524,
      "num_tokens": 19158131.0,
      "step": 8745
    },
    {
      "entropy": 5.506332731246948,
      "epoch": 0.8684001587931719,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004933325294019659,
      "loss": 5.2187,
      "mean_token_accuracy": 0.1621017947793007,
      "num_tokens": 19168820.0,
      "step": 8750
    },
    {
      "entropy": 5.590275287628174,
      "epoch": 0.8688963874553394,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004933239651463068,
      "loss": 5.3073,
      "mean_token_accuracy": 0.16791141927242278,
      "num_tokens": 19178845.0,
      "step": 8755
    },
    {
      "entropy": 5.573048782348633,
      "epoch": 0.869392616117507,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004933153954766837,
      "loss": 5.2843,
      "mean_token_accuracy": 0.1613463819026947,
      "num_tokens": 19190830.0,
      "step": 8760
    },
    {
      "entropy": 5.404928827285767,
      "epoch": 0.8698888447796744,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004933068203933092,
      "loss": 5.144,
      "mean_token_accuracy": 0.1659684345126152,
      "num_tokens": 19202056.0,
      "step": 8765
    },
    {
      "entropy": 5.511019325256347,
      "epoch": 0.870385073441842,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000493298239896396,
      "loss": 5.3581,
      "mean_token_accuracy": 0.15750947892665862,
      "num_tokens": 19213497.0,
      "step": 8770
    },
    {
      "entropy": 5.633516693115235,
      "epoch": 0.8708813021040095,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004932896539861568,
      "loss": 5.3812,
      "mean_token_accuracy": 0.15700077563524245,
      "num_tokens": 19226103.0,
      "step": 8775
    },
    {
      "entropy": 5.575097703933716,
      "epoch": 0.8713775307661771,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004932810626628047,
      "loss": 5.4008,
      "mean_token_accuracy": 0.15823982506990433,
      "num_tokens": 19237297.0,
      "step": 8780
    },
    {
      "entropy": 5.556234884262085,
      "epoch": 0.8718737594283446,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004932724659265526,
      "loss": 5.3055,
      "mean_token_accuracy": 0.15963849276304246,
      "num_tokens": 19247537.0,
      "step": 8785
    },
    {
      "entropy": 5.48999514579773,
      "epoch": 0.8723699880905121,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004932638637776134,
      "loss": 5.2464,
      "mean_token_accuracy": 0.15756665170192719,
      "num_tokens": 19260465.0,
      "step": 8790
    },
    {
      "entropy": 5.549872970581054,
      "epoch": 0.8728662167526796,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000493255256216201,
      "loss": 5.3144,
      "mean_token_accuracy": 0.15558967739343643,
      "num_tokens": 19270534.0,
      "step": 8795
    },
    {
      "entropy": 5.5368547439575195,
      "epoch": 0.8733624454148472,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004932466432425284,
      "loss": 5.328,
      "mean_token_accuracy": 0.1593007780611515,
      "num_tokens": 19281551.0,
      "step": 8800
    },
    {
      "entropy": 5.53004264831543,
      "epoch": 0.8738586740770147,
      "grad_norm": 1.0,
      "learning_rate": 0.0004932380248568094,
      "loss": 5.2722,
      "mean_token_accuracy": 0.1617216259241104,
      "num_tokens": 19293300.0,
      "step": 8805
    },
    {
      "entropy": 5.522213697433472,
      "epoch": 0.8743549027391823,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004932294010592575,
      "loss": 5.2088,
      "mean_token_accuracy": 0.1621657654643059,
      "num_tokens": 19302701.0,
      "step": 8810
    },
    {
      "entropy": 5.561089992523193,
      "epoch": 0.8748511314013497,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004932207718500868,
      "loss": 5.3038,
      "mean_token_accuracy": 0.1600021854043007,
      "num_tokens": 19313024.0,
      "step": 8815
    },
    {
      "entropy": 5.554006576538086,
      "epoch": 0.8753473600635173,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004932121372295112,
      "loss": 5.3208,
      "mean_token_accuracy": 0.15621675699949264,
      "num_tokens": 19324014.0,
      "step": 8820
    },
    {
      "entropy": 5.5339195251464846,
      "epoch": 0.8758435887256848,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004932034971977448,
      "loss": 5.3327,
      "mean_token_accuracy": 0.15784318298101424,
      "num_tokens": 19334226.0,
      "step": 8825
    },
    {
      "entropy": 5.585725498199463,
      "epoch": 0.8763398173878523,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004931948517550019,
      "loss": 5.3665,
      "mean_token_accuracy": 0.15267175287008286,
      "num_tokens": 19344363.0,
      "step": 8830
    },
    {
      "entropy": 5.573243904113769,
      "epoch": 0.8768360460500199,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004931862009014968,
      "loss": 5.3827,
      "mean_token_accuracy": 0.1558567300438881,
      "num_tokens": 19354915.0,
      "step": 8835
    },
    {
      "entropy": 5.623068523406983,
      "epoch": 0.8773322747121873,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000493177544637444,
      "loss": 5.3872,
      "mean_token_accuracy": 0.15675173848867416,
      "num_tokens": 19365570.0,
      "step": 8840
    },
    {
      "entropy": 5.614332580566407,
      "epoch": 0.8778285033743549,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004931688829630582,
      "loss": 5.3009,
      "mean_token_accuracy": 0.16717157363891602,
      "num_tokens": 19375731.0,
      "step": 8845
    },
    {
      "entropy": 5.511403656005859,
      "epoch": 0.8783247320365224,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004931602158785544,
      "loss": 5.2702,
      "mean_token_accuracy": 0.16010398268699647,
      "num_tokens": 19386248.0,
      "step": 8850
    },
    {
      "entropy": 5.497075843811035,
      "epoch": 0.87882096069869,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004931515433841472,
      "loss": 5.2717,
      "mean_token_accuracy": 0.16560742855072022,
      "num_tokens": 19397593.0,
      "step": 8855
    },
    {
      "entropy": 5.558141374588013,
      "epoch": 0.8793171893608575,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004931428654800518,
      "loss": 5.3527,
      "mean_token_accuracy": 0.15407872051000596,
      "num_tokens": 19409701.0,
      "step": 8860
    },
    {
      "entropy": 5.589240264892578,
      "epoch": 0.879813418023025,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004931341821664834,
      "loss": 5.3276,
      "mean_token_accuracy": 0.15600220263004302,
      "num_tokens": 19420280.0,
      "step": 8865
    },
    {
      "entropy": 5.584590053558349,
      "epoch": 0.8803096466851925,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004931254934436573,
      "loss": 5.314,
      "mean_token_accuracy": 0.15992545783519746,
      "num_tokens": 19432208.0,
      "step": 8870
    },
    {
      "entropy": 5.6468058109283445,
      "epoch": 0.8808058753473601,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004931167993117891,
      "loss": 5.4488,
      "mean_token_accuracy": 0.15389225333929063,
      "num_tokens": 19443333.0,
      "step": 8875
    },
    {
      "entropy": 5.497191619873047,
      "epoch": 0.8813021040095276,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004931080997710942,
      "loss": 5.1827,
      "mean_token_accuracy": 0.16432267278432847,
      "num_tokens": 19452867.0,
      "step": 8880
    },
    {
      "entropy": 5.483661651611328,
      "epoch": 0.8817983326716952,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004930993948217884,
      "loss": 5.2241,
      "mean_token_accuracy": 0.1687967821955681,
      "num_tokens": 19463021.0,
      "step": 8885
    },
    {
      "entropy": 5.5666601181030275,
      "epoch": 0.8822945613338626,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004930906844640876,
      "loss": 5.3581,
      "mean_token_accuracy": 0.1564721181988716,
      "num_tokens": 19474092.0,
      "step": 8890
    },
    {
      "entropy": 5.524877643585205,
      "epoch": 0.8827907899960302,
      "grad_norm": 0.875,
      "learning_rate": 0.0004930819686982077,
      "loss": 5.3319,
      "mean_token_accuracy": 0.1533910483121872,
      "num_tokens": 19486035.0,
      "step": 8895
    },
    {
      "entropy": 5.497438240051269,
      "epoch": 0.8832870186581977,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000493073247524365,
      "loss": 5.2632,
      "mean_token_accuracy": 0.16072648912668228,
      "num_tokens": 19497484.0,
      "step": 8900
    },
    {
      "entropy": 5.568736743927002,
      "epoch": 0.8837832473203652,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004930645209427756,
      "loss": 5.3694,
      "mean_token_accuracy": 0.15094278901815414,
      "num_tokens": 19508463.0,
      "step": 8905
    },
    {
      "entropy": 5.570272445678711,
      "epoch": 0.8842794759825328,
      "grad_norm": 0.84375,
      "learning_rate": 0.000493055788953656,
      "loss": 5.3055,
      "mean_token_accuracy": 0.16139017641544343,
      "num_tokens": 19520260.0,
      "step": 8910
    },
    {
      "entropy": 5.536235666275024,
      "epoch": 0.8847757046447002,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004930470515572227,
      "loss": 5.2565,
      "mean_token_accuracy": 0.16148921102285385,
      "num_tokens": 19531621.0,
      "step": 8915
    },
    {
      "entropy": 5.542524147033691,
      "epoch": 0.8852719333068678,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004930383087536923,
      "loss": 5.2437,
      "mean_token_accuracy": 0.15747498124837875,
      "num_tokens": 19543271.0,
      "step": 8920
    },
    {
      "entropy": 5.6115631580352785,
      "epoch": 0.8857681619690353,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004930295605432818,
      "loss": 5.4049,
      "mean_token_accuracy": 0.16077397018671036,
      "num_tokens": 19555178.0,
      "step": 8925
    },
    {
      "entropy": 5.574784278869629,
      "epoch": 0.8862643906312029,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004930208069262079,
      "loss": 5.3257,
      "mean_token_accuracy": 0.16025338172912598,
      "num_tokens": 19565153.0,
      "step": 8930
    },
    {
      "entropy": 5.488508558273315,
      "epoch": 0.8867606192933704,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004930120479026878,
      "loss": 5.2733,
      "mean_token_accuracy": 0.16610471904277802,
      "num_tokens": 19575608.0,
      "step": 8935
    },
    {
      "entropy": 5.554893922805786,
      "epoch": 0.8872568479555379,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004930032834729388,
      "loss": 5.2273,
      "mean_token_accuracy": 0.16715442389249802,
      "num_tokens": 19587556.0,
      "step": 8940
    },
    {
      "entropy": 5.515234899520874,
      "epoch": 0.8877530766177054,
      "grad_norm": 0.890625,
      "learning_rate": 0.000492994513637178,
      "loss": 5.2688,
      "mean_token_accuracy": 0.16076616793870926,
      "num_tokens": 19597341.0,
      "step": 8945
    },
    {
      "entropy": 5.507076930999756,
      "epoch": 0.888249305279873,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004929857383956229,
      "loss": 5.2866,
      "mean_token_accuracy": 0.1587548226118088,
      "num_tokens": 19608431.0,
      "step": 8950
    },
    {
      "entropy": 5.525849533081055,
      "epoch": 0.8887455339420405,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004929769577484914,
      "loss": 5.2795,
      "mean_token_accuracy": 0.16742042303085328,
      "num_tokens": 19619645.0,
      "step": 8955
    },
    {
      "entropy": 5.509834575653076,
      "epoch": 0.8892417626042081,
      "grad_norm": 0.984375,
      "learning_rate": 0.000492968171696001,
      "loss": 5.2425,
      "mean_token_accuracy": 0.16633469015359878,
      "num_tokens": 19630412.0,
      "step": 8960
    },
    {
      "entropy": 5.550708866119384,
      "epoch": 0.8897379912663755,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004929593802383696,
      "loss": 5.2967,
      "mean_token_accuracy": 0.16355745047330855,
      "num_tokens": 19640055.0,
      "step": 8965
    },
    {
      "entropy": 5.504737377166748,
      "epoch": 0.8902342199285431,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004929505833758152,
      "loss": 5.3128,
      "mean_token_accuracy": 0.15629351809620856,
      "num_tokens": 19650814.0,
      "step": 8970
    },
    {
      "entropy": 5.560285472869873,
      "epoch": 0.8907304485907106,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004929417811085561,
      "loss": 5.3086,
      "mean_token_accuracy": 0.15954534113407134,
      "num_tokens": 19663669.0,
      "step": 8975
    },
    {
      "entropy": 5.634855079650879,
      "epoch": 0.8912266772528781,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004929329734368103,
      "loss": 5.3851,
      "mean_token_accuracy": 0.1564209520816803,
      "num_tokens": 19675851.0,
      "step": 8980
    },
    {
      "entropy": 5.549490594863892,
      "epoch": 0.8917229059150457,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004929241603607964,
      "loss": 5.2718,
      "mean_token_accuracy": 0.1691298305988312,
      "num_tokens": 19686613.0,
      "step": 8985
    },
    {
      "entropy": 5.466474437713623,
      "epoch": 0.8922191345772131,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000492915341880733,
      "loss": 5.1992,
      "mean_token_accuracy": 0.16453103274106978,
      "num_tokens": 19696419.0,
      "step": 8990
    },
    {
      "entropy": 5.518602275848389,
      "epoch": 0.8927153632393807,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004929065179968387,
      "loss": 5.2944,
      "mean_token_accuracy": 0.15724054127931594,
      "num_tokens": 19707269.0,
      "step": 8995
    },
    {
      "entropy": 5.605218172073364,
      "epoch": 0.8932115919015482,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004928976887093321,
      "loss": 5.341,
      "mean_token_accuracy": 0.15894262343645096,
      "num_tokens": 19716099.0,
      "step": 9000
    },
    {
      "epoch": 0.8932115919015482,
      "eval_entropy": 5.3559456596637345,
      "eval_loss": 5.37523889541626,
      "eval_mean_token_accuracy": 0.16191325234025752,
      "eval_num_tokens": 19716099.0,
      "eval_runtime": 20.6605,
      "eval_samples_per_second": 1629.629,
      "eval_steps_per_second": 203.722,
      "step": 9000
    },
    {
      "entropy": 5.54390869140625,
      "epoch": 0.8937078205637158,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004928888540184325,
      "loss": 5.3313,
      "mean_token_accuracy": 0.15764046013355254,
      "num_tokens": 19726632.0,
      "step": 9005
    },
    {
      "entropy": 5.51641879081726,
      "epoch": 0.8942040492258833,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004928800139243587,
      "loss": 5.2468,
      "mean_token_accuracy": 0.16316547393798828,
      "num_tokens": 19737929.0,
      "step": 9010
    },
    {
      "entropy": 5.5370947360992435,
      "epoch": 0.8947002778880508,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004928711684273302,
      "loss": 5.2577,
      "mean_token_accuracy": 0.16472524553537368,
      "num_tokens": 19748191.0,
      "step": 9015
    },
    {
      "entropy": 5.542944002151489,
      "epoch": 0.8951965065502183,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004928623175275661,
      "loss": 5.3148,
      "mean_token_accuracy": 0.16453443765640258,
      "num_tokens": 19758892.0,
      "step": 9020
    },
    {
      "entropy": 5.499523258209228,
      "epoch": 0.8956927352123859,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004928534612252862,
      "loss": 5.1886,
      "mean_token_accuracy": 0.16720210164785385,
      "num_tokens": 19769120.0,
      "step": 9025
    },
    {
      "entropy": 5.499724817276001,
      "epoch": 0.8961889638745534,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004928445995207095,
      "loss": 5.2481,
      "mean_token_accuracy": 0.16502394229173661,
      "num_tokens": 19779484.0,
      "step": 9030
    },
    {
      "entropy": 5.5149274349212645,
      "epoch": 0.896685192536721,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004928357324140564,
      "loss": 5.2963,
      "mean_token_accuracy": 0.16460447013378143,
      "num_tokens": 19790438.0,
      "step": 9035
    },
    {
      "entropy": 5.583444356918335,
      "epoch": 0.8971814211988884,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004928268599055465,
      "loss": 5.37,
      "mean_token_accuracy": 0.15877770036458969,
      "num_tokens": 19801646.0,
      "step": 9040
    },
    {
      "entropy": 5.560933637619018,
      "epoch": 0.897677649861056,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004928179819953998,
      "loss": 5.2606,
      "mean_token_accuracy": 0.16284436732530594,
      "num_tokens": 19812887.0,
      "step": 9045
    },
    {
      "entropy": 5.540252304077148,
      "epoch": 0.8981738785232235,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004928090986838366,
      "loss": 5.3149,
      "mean_token_accuracy": 0.1579166457056999,
      "num_tokens": 19822938.0,
      "step": 9050
    },
    {
      "entropy": 5.507092380523682,
      "epoch": 0.898670107185391,
      "grad_norm": 1.078125,
      "learning_rate": 0.000492800209971077,
      "loss": 5.271,
      "mean_token_accuracy": 0.16255158483982085,
      "num_tokens": 19833988.0,
      "step": 9055
    },
    {
      "entropy": 5.596437597274781,
      "epoch": 0.8991663358475586,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004927913158573416,
      "loss": 5.3345,
      "mean_token_accuracy": 0.15781496465206146,
      "num_tokens": 19845112.0,
      "step": 9060
    },
    {
      "entropy": 5.525338649749756,
      "epoch": 0.899662564509726,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004927824163428508,
      "loss": 5.2947,
      "mean_token_accuracy": 0.1584398239850998,
      "num_tokens": 19856304.0,
      "step": 9065
    },
    {
      "entropy": 5.5639386653900145,
      "epoch": 0.9001587931718936,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004927735114278254,
      "loss": 5.2946,
      "mean_token_accuracy": 0.16316746026277543,
      "num_tokens": 19866776.0,
      "step": 9070
    },
    {
      "entropy": 5.566666889190674,
      "epoch": 0.9006550218340611,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004927646011124861,
      "loss": 5.2598,
      "mean_token_accuracy": 0.1567187190055847,
      "num_tokens": 19876392.0,
      "step": 9075
    },
    {
      "entropy": 5.559143972396851,
      "epoch": 0.9011512504962287,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004927556853970539,
      "loss": 5.2991,
      "mean_token_accuracy": 0.15610761046409607,
      "num_tokens": 19887362.0,
      "step": 9080
    },
    {
      "entropy": 5.508514881134033,
      "epoch": 0.9016474791583962,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004927467642817499,
      "loss": 5.277,
      "mean_token_accuracy": 0.16156834959983826,
      "num_tokens": 19898989.0,
      "step": 9085
    },
    {
      "entropy": 5.551664209365844,
      "epoch": 0.9021437078205637,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004927378377667955,
      "loss": 5.2741,
      "mean_token_accuracy": 0.15657708197832107,
      "num_tokens": 19910112.0,
      "step": 9090
    },
    {
      "entropy": 5.606841802597046,
      "epoch": 0.9026399364827312,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000492728905852412,
      "loss": 5.3956,
      "mean_token_accuracy": 0.15744143724441528,
      "num_tokens": 19922601.0,
      "step": 9095
    },
    {
      "entropy": 5.5513715744018555,
      "epoch": 0.9031361651448988,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004927199685388205,
      "loss": 5.303,
      "mean_token_accuracy": 0.1631772145628929,
      "num_tokens": 19934153.0,
      "step": 9100
    },
    {
      "entropy": 5.535596609115601,
      "epoch": 0.9036323938070663,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000492711025826243,
      "loss": 5.2764,
      "mean_token_accuracy": 0.16301968693733215,
      "num_tokens": 19944572.0,
      "step": 9105
    },
    {
      "entropy": 5.546830320358277,
      "epoch": 0.9041286224692339,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004927020777149014,
      "loss": 5.3893,
      "mean_token_accuracy": 0.15813228785991668,
      "num_tokens": 19956431.0,
      "step": 9110
    },
    {
      "entropy": 5.589463758468628,
      "epoch": 0.9046248511314013,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004926931242050171,
      "loss": 5.3052,
      "mean_token_accuracy": 0.15729486644268037,
      "num_tokens": 19968118.0,
      "step": 9115
    },
    {
      "entropy": 5.540275049209595,
      "epoch": 0.9051210797935689,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004926841652968125,
      "loss": 5.3725,
      "mean_token_accuracy": 0.15925574228167533,
      "num_tokens": 19979545.0,
      "step": 9120
    },
    {
      "entropy": 5.543299961090088,
      "epoch": 0.9056173084557364,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004926752009905099,
      "loss": 5.3314,
      "mean_token_accuracy": 0.15359800308942795,
      "num_tokens": 19989832.0,
      "step": 9125
    },
    {
      "entropy": 5.542871141433716,
      "epoch": 0.9061135371179039,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004926662312863311,
      "loss": 5.3397,
      "mean_token_accuracy": 0.15784236341714858,
      "num_tokens": 20000935.0,
      "step": 9130
    },
    {
      "entropy": 5.546960735321045,
      "epoch": 0.9066097657800715,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004926572561844988,
      "loss": 5.3435,
      "mean_token_accuracy": 0.15512987971305847,
      "num_tokens": 20011485.0,
      "step": 9135
    },
    {
      "entropy": 5.513018083572388,
      "epoch": 0.9071059944422389,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004926482756852356,
      "loss": 5.1953,
      "mean_token_accuracy": 0.16653136759996415,
      "num_tokens": 20023043.0,
      "step": 9140
    },
    {
      "entropy": 5.572578573226929,
      "epoch": 0.9076022231044065,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004926392897887642,
      "loss": 5.2971,
      "mean_token_accuracy": 0.16815751641988755,
      "num_tokens": 20032437.0,
      "step": 9145
    },
    {
      "entropy": 5.553290939331054,
      "epoch": 0.908098451766574,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004926302984953073,
      "loss": 5.3356,
      "mean_token_accuracy": 0.15907517820596695,
      "num_tokens": 20044662.0,
      "step": 9150
    },
    {
      "entropy": 5.527995777130127,
      "epoch": 0.9085946804287416,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004926213018050881,
      "loss": 5.3433,
      "mean_token_accuracy": 0.1597750887274742,
      "num_tokens": 20055924.0,
      "step": 9155
    },
    {
      "entropy": 5.503418397903443,
      "epoch": 0.9090909090909091,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004926122997183295,
      "loss": 5.2934,
      "mean_token_accuracy": 0.15803872793912888,
      "num_tokens": 20067909.0,
      "step": 9160
    },
    {
      "entropy": 5.570617198944092,
      "epoch": 0.9095871377530766,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004926032922352546,
      "loss": 5.2679,
      "mean_token_accuracy": 0.16587397903203965,
      "num_tokens": 20079333.0,
      "step": 9165
    },
    {
      "entropy": 5.548792457580566,
      "epoch": 0.9100833664152441,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004925942793560872,
      "loss": 5.2517,
      "mean_token_accuracy": 0.1625090569257736,
      "num_tokens": 20089741.0,
      "step": 9170
    },
    {
      "entropy": 5.4624285221099855,
      "epoch": 0.9105795950774117,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004925852610810504,
      "loss": 5.2682,
      "mean_token_accuracy": 0.16062215268611907,
      "num_tokens": 20100267.0,
      "step": 9175
    },
    {
      "entropy": 5.504045677185059,
      "epoch": 0.9110758237395792,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000492576237410368,
      "loss": 5.2625,
      "mean_token_accuracy": 0.16514603048563004,
      "num_tokens": 20111964.0,
      "step": 9180
    },
    {
      "entropy": 5.567818880081177,
      "epoch": 0.9115720524017468,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004925672083442638,
      "loss": 5.3108,
      "mean_token_accuracy": 0.1528840556740761,
      "num_tokens": 20123132.0,
      "step": 9185
    },
    {
      "entropy": 5.469148397445679,
      "epoch": 0.9120682810639142,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004925581738829616,
      "loss": 5.2147,
      "mean_token_accuracy": 0.16623493880033494,
      "num_tokens": 20134050.0,
      "step": 9190
    },
    {
      "entropy": 5.490215921401978,
      "epoch": 0.9125645097260818,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004925491340266855,
      "loss": 5.2723,
      "mean_token_accuracy": 0.16605161279439926,
      "num_tokens": 20143814.0,
      "step": 9195
    },
    {
      "entropy": 5.5671141147613525,
      "epoch": 0.9130607383882493,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004925400887756596,
      "loss": 5.3543,
      "mean_token_accuracy": 0.1668687030673027,
      "num_tokens": 20157647.0,
      "step": 9200
    },
    {
      "entropy": 5.553544712066651,
      "epoch": 0.9135569670504168,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004925310381301083,
      "loss": 5.3219,
      "mean_token_accuracy": 0.16231947541236877,
      "num_tokens": 20168366.0,
      "step": 9205
    },
    {
      "entropy": 5.518465995788574,
      "epoch": 0.9140531957125844,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000492521982090256,
      "loss": 5.3712,
      "mean_token_accuracy": 0.1649198353290558,
      "num_tokens": 20179362.0,
      "step": 9210
    },
    {
      "entropy": 5.496027278900146,
      "epoch": 0.9145494243747518,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004925129206563273,
      "loss": 5.2507,
      "mean_token_accuracy": 0.16224647611379622,
      "num_tokens": 20190320.0,
      "step": 9215
    },
    {
      "entropy": 5.514825487136841,
      "epoch": 0.9150456530369194,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004925038538285467,
      "loss": 5.2776,
      "mean_token_accuracy": 0.1612439125776291,
      "num_tokens": 20201114.0,
      "step": 9220
    },
    {
      "entropy": 5.568852090835572,
      "epoch": 0.9155418816990869,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004924947816071394,
      "loss": 5.3015,
      "mean_token_accuracy": 0.16260259449481965,
      "num_tokens": 20213351.0,
      "step": 9225
    },
    {
      "entropy": 5.4828455448150635,
      "epoch": 0.9160381103612545,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00049248570399233,
      "loss": 5.1966,
      "mean_token_accuracy": 0.167659492790699,
      "num_tokens": 20223376.0,
      "step": 9230
    },
    {
      "entropy": 5.5454470157623295,
      "epoch": 0.916534339023422,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000492476620984344,
      "loss": 5.3408,
      "mean_token_accuracy": 0.15844542682170867,
      "num_tokens": 20234248.0,
      "step": 9235
    },
    {
      "entropy": 5.493923091888428,
      "epoch": 0.9170305676855895,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004924675325834063,
      "loss": 5.2665,
      "mean_token_accuracy": 0.16199109256267546,
      "num_tokens": 20245823.0,
      "step": 9240
    },
    {
      "entropy": 5.567879724502563,
      "epoch": 0.917526796347757,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004924584387897424,
      "loss": 5.3182,
      "mean_token_accuracy": 0.15727784782648085,
      "num_tokens": 20257003.0,
      "step": 9245
    },
    {
      "entropy": 5.5870460033416744,
      "epoch": 0.9180230250099246,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004924493396035777,
      "loss": 5.4568,
      "mean_token_accuracy": 0.15009055137634278,
      "num_tokens": 20268500.0,
      "step": 9250
    },
    {
      "entropy": 5.580854177474976,
      "epoch": 0.9185192536720921,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000492440235025138,
      "loss": 5.3211,
      "mean_token_accuracy": 0.16316964626312255,
      "num_tokens": 20279301.0,
      "step": 9255
    },
    {
      "entropy": 5.668870878219605,
      "epoch": 0.9190154823342597,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924311250546491,
      "loss": 5.4184,
      "mean_token_accuracy": 0.15555869787931442,
      "num_tokens": 20291263.0,
      "step": 9260
    },
    {
      "entropy": 5.546143102645874,
      "epoch": 0.9195117109964271,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004924220096923367,
      "loss": 5.2915,
      "mean_token_accuracy": 0.16107711046934128,
      "num_tokens": 20301554.0,
      "step": 9265
    },
    {
      "entropy": 5.597596216201782,
      "epoch": 0.9200079396585947,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004924128889384271,
      "loss": 5.3434,
      "mean_token_accuracy": 0.1540423795580864,
      "num_tokens": 20314299.0,
      "step": 9270
    },
    {
      "entropy": 5.5163649559021,
      "epoch": 0.9205041683207622,
      "grad_norm": 0.875,
      "learning_rate": 0.0004924037627931463,
      "loss": 5.3096,
      "mean_token_accuracy": 0.1569603279232979,
      "num_tokens": 20324595.0,
      "step": 9275
    },
    {
      "entropy": 5.533094978332519,
      "epoch": 0.9210003969829297,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004923946312567206,
      "loss": 5.2775,
      "mean_token_accuracy": 0.15990834385156633,
      "num_tokens": 20335744.0,
      "step": 9280
    },
    {
      "entropy": 5.477016162872315,
      "epoch": 0.9214966256450973,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004923854943293765,
      "loss": 5.1769,
      "mean_token_accuracy": 0.16062400490045547,
      "num_tokens": 20346072.0,
      "step": 9285
    },
    {
      "entropy": 5.51767783164978,
      "epoch": 0.9219928543072647,
      "grad_norm": 0.78125,
      "learning_rate": 0.0004923763520113406,
      "loss": 5.2671,
      "mean_token_accuracy": 0.17054644227027893,
      "num_tokens": 20357747.0,
      "step": 9290
    },
    {
      "entropy": 5.589597606658936,
      "epoch": 0.9224890829694323,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004923672043028395,
      "loss": 5.2477,
      "mean_token_accuracy": 0.1620402678847313,
      "num_tokens": 20368838.0,
      "step": 9295
    },
    {
      "entropy": 5.542013216018677,
      "epoch": 0.9229853116315998,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004923580512041001,
      "loss": 5.3211,
      "mean_token_accuracy": 0.15841785520315171,
      "num_tokens": 20380844.0,
      "step": 9300
    },
    {
      "entropy": 5.5596929550170895,
      "epoch": 0.9234815402937674,
      "grad_norm": 0.875,
      "learning_rate": 0.0004923488927153495,
      "loss": 5.3283,
      "mean_token_accuracy": 0.15808232724666596,
      "num_tokens": 20391739.0,
      "step": 9305
    },
    {
      "entropy": 5.5594805717468265,
      "epoch": 0.9239777689559349,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004923397288368146,
      "loss": 5.2179,
      "mean_token_accuracy": 0.17128596156835557,
      "num_tokens": 20402571.0,
      "step": 9310
    },
    {
      "entropy": 5.515147495269775,
      "epoch": 0.9244739976181024,
      "grad_norm": 1.0,
      "learning_rate": 0.0004923305595687229,
      "loss": 5.2554,
      "mean_token_accuracy": 0.16113622933626176,
      "num_tokens": 20414766.0,
      "step": 9315
    },
    {
      "entropy": 5.481299829483032,
      "epoch": 0.9249702262802699,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004923213849113015,
      "loss": 5.2268,
      "mean_token_accuracy": 0.16429373770952224,
      "num_tokens": 20425508.0,
      "step": 9320
    },
    {
      "entropy": 5.472104215621949,
      "epoch": 0.9254664549424375,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004923122048647781,
      "loss": 5.2209,
      "mean_token_accuracy": 0.1635184809565544,
      "num_tokens": 20436128.0,
      "step": 9325
    },
    {
      "entropy": 5.50439944267273,
      "epoch": 0.925962683604605,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004923030194293802,
      "loss": 5.3407,
      "mean_token_accuracy": 0.15802000164985658,
      "num_tokens": 20447212.0,
      "step": 9330
    },
    {
      "entropy": 5.541323184967041,
      "epoch": 0.9264589122667726,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004922938286053358,
      "loss": 5.2536,
      "mean_token_accuracy": 0.16571350246667862,
      "num_tokens": 20457770.0,
      "step": 9335
    },
    {
      "entropy": 5.534483146667481,
      "epoch": 0.92695514092894,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004922846323928726,
      "loss": 5.3223,
      "mean_token_accuracy": 0.15829780399799348,
      "num_tokens": 20468172.0,
      "step": 9340
    },
    {
      "entropy": 5.518729972839355,
      "epoch": 0.9274513695911076,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004922754307922188,
      "loss": 5.2452,
      "mean_token_accuracy": 0.16664718389511107,
      "num_tokens": 20479429.0,
      "step": 9345
    },
    {
      "entropy": 5.532600736618042,
      "epoch": 0.9279475982532751,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004922662238036024,
      "loss": 5.2169,
      "mean_token_accuracy": 0.16647566258907318,
      "num_tokens": 20490373.0,
      "step": 9350
    },
    {
      "entropy": 5.591423892974854,
      "epoch": 0.9284438269154426,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004922570114272518,
      "loss": 5.3467,
      "mean_token_accuracy": 0.16103961020708085,
      "num_tokens": 20500002.0,
      "step": 9355
    },
    {
      "entropy": 5.58456506729126,
      "epoch": 0.9289400555776102,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004922477936633955,
      "loss": 5.3858,
      "mean_token_accuracy": 0.1511903539299965,
      "num_tokens": 20512990.0,
      "step": 9360
    },
    {
      "entropy": 5.487053728103637,
      "epoch": 0.9294362842397776,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004922385705122621,
      "loss": 5.2877,
      "mean_token_accuracy": 0.16572093963623047,
      "num_tokens": 20522979.0,
      "step": 9365
    },
    {
      "entropy": 5.482340526580811,
      "epoch": 0.9299325129019452,
      "grad_norm": 0.875,
      "learning_rate": 0.0004922293419740802,
      "loss": 5.1863,
      "mean_token_accuracy": 0.16835395693778993,
      "num_tokens": 20534620.0,
      "step": 9370
    },
    {
      "entropy": 5.542438554763794,
      "epoch": 0.9304287415641127,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004922201080490787,
      "loss": 5.2794,
      "mean_token_accuracy": 0.16248394697904586,
      "num_tokens": 20545526.0,
      "step": 9375
    },
    {
      "entropy": 5.605936765670776,
      "epoch": 0.9309249702262803,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004922108687374867,
      "loss": 5.4447,
      "mean_token_accuracy": 0.15707781314849853,
      "num_tokens": 20556665.0,
      "step": 9380
    },
    {
      "entropy": 5.571640491485596,
      "epoch": 0.9314211988884478,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000492201624039533,
      "loss": 5.2937,
      "mean_token_accuracy": 0.1559583067893982,
      "num_tokens": 20566054.0,
      "step": 9385
    },
    {
      "entropy": 5.631577730178833,
      "epoch": 0.9319174275506154,
      "grad_norm": 0.875,
      "learning_rate": 0.0004921923739554472,
      "loss": 5.355,
      "mean_token_accuracy": 0.15491781681776046,
      "num_tokens": 20577922.0,
      "step": 9390
    },
    {
      "entropy": 5.471510410308838,
      "epoch": 0.9324136562127828,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004921831184854585,
      "loss": 5.1735,
      "mean_token_accuracy": 0.16721322238445283,
      "num_tokens": 20588473.0,
      "step": 9395
    },
    {
      "entropy": 5.51868200302124,
      "epoch": 0.9329098848749504,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004921738576297965,
      "loss": 5.3654,
      "mean_token_accuracy": 0.15697364658117294,
      "num_tokens": 20599371.0,
      "step": 9400
    },
    {
      "entropy": 5.505065202713013,
      "epoch": 0.9334061135371179,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004921645913886907,
      "loss": 5.2431,
      "mean_token_accuracy": 0.15694219917058944,
      "num_tokens": 20610066.0,
      "step": 9405
    },
    {
      "entropy": 5.560163259506226,
      "epoch": 0.9339023421992855,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004921553197623711,
      "loss": 5.3424,
      "mean_token_accuracy": 0.1578836128115654,
      "num_tokens": 20621804.0,
      "step": 9410
    },
    {
      "entropy": 5.563719654083252,
      "epoch": 0.934398570861453,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004921460427510674,
      "loss": 5.2359,
      "mean_token_accuracy": 0.16877928823232652,
      "num_tokens": 20632706.0,
      "step": 9415
    },
    {
      "entropy": 5.520000648498535,
      "epoch": 0.9348947995236205,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004921367603550098,
      "loss": 5.3352,
      "mean_token_accuracy": 0.15229426324367523,
      "num_tokens": 20643303.0,
      "step": 9420
    },
    {
      "entropy": 5.530603313446045,
      "epoch": 0.935391028185788,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004921274725744283,
      "loss": 5.2895,
      "mean_token_accuracy": 0.1678344264626503,
      "num_tokens": 20653681.0,
      "step": 9425
    },
    {
      "entropy": 5.5783344268798825,
      "epoch": 0.9358872568479555,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004921181794095536,
      "loss": 5.3101,
      "mean_token_accuracy": 0.1571393668651581,
      "num_tokens": 20665599.0,
      "step": 9430
    },
    {
      "entropy": 5.48060564994812,
      "epoch": 0.9363834855101231,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004921088808606157,
      "loss": 5.1938,
      "mean_token_accuracy": 0.1647046461701393,
      "num_tokens": 20675726.0,
      "step": 9435
    },
    {
      "entropy": 5.522446250915527,
      "epoch": 0.9368797141722905,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004920995769278454,
      "loss": 5.3764,
      "mean_token_accuracy": 0.1545850783586502,
      "num_tokens": 20688083.0,
      "step": 9440
    },
    {
      "entropy": 5.562725925445557,
      "epoch": 0.9373759428344581,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004920902676114734,
      "loss": 5.2386,
      "mean_token_accuracy": 0.16600545197725297,
      "num_tokens": 20697642.0,
      "step": 9445
    },
    {
      "entropy": 5.575511169433594,
      "epoch": 0.9378721714966256,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004920809529117306,
      "loss": 5.3388,
      "mean_token_accuracy": 0.15686559230089187,
      "num_tokens": 20709172.0,
      "step": 9450
    },
    {
      "entropy": 5.600897645950317,
      "epoch": 0.9383684001587932,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004920716328288479,
      "loss": 5.3896,
      "mean_token_accuracy": 0.1551354184746742,
      "num_tokens": 20720563.0,
      "step": 9455
    },
    {
      "entropy": 5.54432897567749,
      "epoch": 0.9388646288209607,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004920623073630566,
      "loss": 5.2686,
      "mean_token_accuracy": 0.1644684612751007,
      "num_tokens": 20730943.0,
      "step": 9460
    },
    {
      "entropy": 5.539048767089843,
      "epoch": 0.9393608574831283,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004920529765145876,
      "loss": 5.2408,
      "mean_token_accuracy": 0.16011967062950133,
      "num_tokens": 20740903.0,
      "step": 9465
    },
    {
      "entropy": 5.531922006607056,
      "epoch": 0.9398570861452957,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004920436402836726,
      "loss": 5.3323,
      "mean_token_accuracy": 0.16579257398843766,
      "num_tokens": 20750580.0,
      "step": 9470
    },
    {
      "entropy": 5.530846405029297,
      "epoch": 0.9403533148074633,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000492034298670543,
      "loss": 5.2539,
      "mean_token_accuracy": 0.16114427000284196,
      "num_tokens": 20761357.0,
      "step": 9475
    },
    {
      "entropy": 5.625053882598877,
      "epoch": 0.9408495434696308,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004920249516754304,
      "loss": 5.4096,
      "mean_token_accuracy": 0.1506986804306507,
      "num_tokens": 20773273.0,
      "step": 9480
    },
    {
      "entropy": 5.5660483837127686,
      "epoch": 0.9413457721317984,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004920155992985667,
      "loss": 5.2401,
      "mean_token_accuracy": 0.16067681461572647,
      "num_tokens": 20784290.0,
      "step": 9485
    },
    {
      "entropy": 5.531092119216919,
      "epoch": 0.9418420007939659,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004920062415401837,
      "loss": 5.2741,
      "mean_token_accuracy": 0.16394226998090744,
      "num_tokens": 20794764.0,
      "step": 9490
    },
    {
      "entropy": 5.564286994934082,
      "epoch": 0.9423382294561334,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004919968784005136,
      "loss": 5.3155,
      "mean_token_accuracy": 0.15846064537763596,
      "num_tokens": 20804863.0,
      "step": 9495
    },
    {
      "entropy": 5.6152464866638185,
      "epoch": 0.9428344581183009,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004919875098797886,
      "loss": 5.4012,
      "mean_token_accuracy": 0.15524280518293382,
      "num_tokens": 20816445.0,
      "step": 9500
    },
    {
      "entropy": 5.557435178756714,
      "epoch": 0.9433306867804684,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004919781359782407,
      "loss": 5.2244,
      "mean_token_accuracy": 0.16715539544820784,
      "num_tokens": 20827141.0,
      "step": 9505
    },
    {
      "entropy": 5.526000022888184,
      "epoch": 0.943826915442636,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004919687566961027,
      "loss": 5.2707,
      "mean_token_accuracy": 0.15752107948064803,
      "num_tokens": 20838291.0,
      "step": 9510
    },
    {
      "entropy": 5.579127216339112,
      "epoch": 0.9443231441048034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000491959372033607,
      "loss": 5.2776,
      "mean_token_accuracy": 0.16056058555841446,
      "num_tokens": 20848839.0,
      "step": 9515
    },
    {
      "entropy": 5.5651250839233395,
      "epoch": 0.944819372766971,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004919499819909865,
      "loss": 5.2558,
      "mean_token_accuracy": 0.16248184889554979,
      "num_tokens": 20859961.0,
      "step": 9520
    },
    {
      "entropy": 5.481574821472168,
      "epoch": 0.9453156014291385,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004919405865684738,
      "loss": 5.2034,
      "mean_token_accuracy": 0.16611860692501068,
      "num_tokens": 20869720.0,
      "step": 9525
    },
    {
      "entropy": 5.507126235961914,
      "epoch": 0.9458118300913061,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000491931185766302,
      "loss": 5.3435,
      "mean_token_accuracy": 0.16365325152873994,
      "num_tokens": 20880513.0,
      "step": 9530
    },
    {
      "entropy": 5.566145992279052,
      "epoch": 0.9463080587534736,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004919217795847044,
      "loss": 5.2656,
      "mean_token_accuracy": 0.16495689749717712,
      "num_tokens": 20892456.0,
      "step": 9535
    },
    {
      "entropy": 5.534870147705078,
      "epoch": 0.9468042874156412,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004919123680239139,
      "loss": 5.2722,
      "mean_token_accuracy": 0.16199540197849274,
      "num_tokens": 20904500.0,
      "step": 9540
    },
    {
      "entropy": 5.5694770336151125,
      "epoch": 0.9473005160778086,
      "grad_norm": 0.875,
      "learning_rate": 0.0004919029510841641,
      "loss": 5.2541,
      "mean_token_accuracy": 0.16894175112247467,
      "num_tokens": 20915133.0,
      "step": 9545
    },
    {
      "entropy": 5.453264045715332,
      "epoch": 0.9477967447399762,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004918935287656887,
      "loss": 5.245,
      "mean_token_accuracy": 0.16683667302131652,
      "num_tokens": 20924779.0,
      "step": 9550
    },
    {
      "entropy": 5.459113740921021,
      "epoch": 0.9482929734021437,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000491884101068721,
      "loss": 5.1905,
      "mean_token_accuracy": 0.1652270406484604,
      "num_tokens": 20935387.0,
      "step": 9555
    },
    {
      "entropy": 5.619988679885864,
      "epoch": 0.9487892020643113,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004918746679934949,
      "loss": 5.3682,
      "mean_token_accuracy": 0.1594923883676529,
      "num_tokens": 20945432.0,
      "step": 9560
    },
    {
      "entropy": 5.542451524734497,
      "epoch": 0.9492854307264788,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004918652295402444,
      "loss": 5.2483,
      "mean_token_accuracy": 0.1673049211502075,
      "num_tokens": 20956777.0,
      "step": 9565
    },
    {
      "entropy": 5.500725793838501,
      "epoch": 0.9497816593886463,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004918557857092036,
      "loss": 5.2856,
      "mean_token_accuracy": 0.15981982946395873,
      "num_tokens": 20966057.0,
      "step": 9570
    },
    {
      "entropy": 5.551764345169067,
      "epoch": 0.9502778880508138,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004918463365006065,
      "loss": 5.3556,
      "mean_token_accuracy": 0.15958352386951447,
      "num_tokens": 20976838.0,
      "step": 9575
    },
    {
      "entropy": 5.602881717681885,
      "epoch": 0.9507741167129813,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004918368819146875,
      "loss": 5.2996,
      "mean_token_accuracy": 0.1609395667910576,
      "num_tokens": 20988553.0,
      "step": 9580
    },
    {
      "entropy": 5.53433780670166,
      "epoch": 0.9512703453751489,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004918274219516811,
      "loss": 5.2791,
      "mean_token_accuracy": 0.1604821041226387,
      "num_tokens": 20998983.0,
      "step": 9585
    },
    {
      "entropy": 5.625037336349488,
      "epoch": 0.9517665740373163,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004918179566118218,
      "loss": 5.3883,
      "mean_token_accuracy": 0.1573731333017349,
      "num_tokens": 21010008.0,
      "step": 9590
    },
    {
      "entropy": 5.598756742477417,
      "epoch": 0.9522628026994839,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004918084858953445,
      "loss": 5.3599,
      "mean_token_accuracy": 0.16161780506372453,
      "num_tokens": 21022826.0,
      "step": 9595
    },
    {
      "entropy": 5.555862092971802,
      "epoch": 0.9527590313616514,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004917990098024839,
      "loss": 5.2834,
      "mean_token_accuracy": 0.15864650160074234,
      "num_tokens": 21033457.0,
      "step": 9600
    },
    {
      "entropy": 5.505966281890869,
      "epoch": 0.953255260023819,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004917895283334749,
      "loss": 5.3013,
      "mean_token_accuracy": 0.16056045591831208,
      "num_tokens": 21044640.0,
      "step": 9605
    },
    {
      "entropy": 5.567455577850342,
      "epoch": 0.9537514886859865,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000491780041488553,
      "loss": 5.2756,
      "mean_token_accuracy": 0.1714605748653412,
      "num_tokens": 21055728.0,
      "step": 9610
    },
    {
      "entropy": 5.550451135635376,
      "epoch": 0.9542477173481541,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004917705492679529,
      "loss": 5.2707,
      "mean_token_accuracy": 0.15992705672979354,
      "num_tokens": 21066057.0,
      "step": 9615
    },
    {
      "entropy": 5.578377771377563,
      "epoch": 0.9547439460103215,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004917610516719104,
      "loss": 5.2892,
      "mean_token_accuracy": 0.15747786164283753,
      "num_tokens": 21075717.0,
      "step": 9620
    },
    {
      "entropy": 5.647128391265869,
      "epoch": 0.9552401746724891,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004917515487006609,
      "loss": 5.3876,
      "mean_token_accuracy": 0.15149791836738585,
      "num_tokens": 21087120.0,
      "step": 9625
    },
    {
      "entropy": 5.595139646530152,
      "epoch": 0.9557364033346566,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00049174204035444,
      "loss": 5.4064,
      "mean_token_accuracy": 0.15330189242959022,
      "num_tokens": 21100110.0,
      "step": 9630
    },
    {
      "entropy": 5.641125917434692,
      "epoch": 0.9562326319968242,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004917325266334836,
      "loss": 5.3082,
      "mean_token_accuracy": 0.1670064240694046,
      "num_tokens": 21110411.0,
      "step": 9635
    },
    {
      "entropy": 5.508846139907837,
      "epoch": 0.9567288606589917,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004917230075380276,
      "loss": 5.2215,
      "mean_token_accuracy": 0.16506332606077195,
      "num_tokens": 21120303.0,
      "step": 9640
    },
    {
      "entropy": 5.52659010887146,
      "epoch": 0.9572250893211592,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004917134830683079,
      "loss": 5.3249,
      "mean_token_accuracy": 0.15908250510692595,
      "num_tokens": 21132312.0,
      "step": 9645
    },
    {
      "entropy": 5.587954521179199,
      "epoch": 0.9577213179833267,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004917039532245608,
      "loss": 5.2954,
      "mean_token_accuracy": 0.16296826899051667,
      "num_tokens": 21142472.0,
      "step": 9650
    },
    {
      "entropy": 5.547664308547974,
      "epoch": 0.9582175466454942,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004916944180070226,
      "loss": 5.3104,
      "mean_token_accuracy": 0.1562706708908081,
      "num_tokens": 21153306.0,
      "step": 9655
    },
    {
      "entropy": 5.620442676544189,
      "epoch": 0.9587137753076618,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004916848774159299,
      "loss": 5.3544,
      "mean_token_accuracy": 0.15180012881755828,
      "num_tokens": 21165606.0,
      "step": 9660
    },
    {
      "entropy": 5.582696485519409,
      "epoch": 0.9592100039698293,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000491675331451519,
      "loss": 5.2715,
      "mean_token_accuracy": 0.16641474664211273,
      "num_tokens": 21176734.0,
      "step": 9665
    },
    {
      "entropy": 5.496678686141967,
      "epoch": 0.9597062326319968,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004916657801140269,
      "loss": 5.2196,
      "mean_token_accuracy": 0.16818380653858184,
      "num_tokens": 21187109.0,
      "step": 9670
    },
    {
      "entropy": 5.493848991394043,
      "epoch": 0.9602024612941643,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004916562234036903,
      "loss": 5.2714,
      "mean_token_accuracy": 0.16576831340789794,
      "num_tokens": 21197281.0,
      "step": 9675
    },
    {
      "entropy": 5.538375568389893,
      "epoch": 0.9606986899563319,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004916466613207462,
      "loss": 5.3115,
      "mean_token_accuracy": 0.15971529334783555,
      "num_tokens": 21208767.0,
      "step": 9680
    },
    {
      "entropy": 5.5474122047424315,
      "epoch": 0.9611949186184994,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004916370938654317,
      "loss": 5.2969,
      "mean_token_accuracy": 0.16287417113780975,
      "num_tokens": 21219328.0,
      "step": 9685
    },
    {
      "entropy": 5.602007246017456,
      "epoch": 0.961691147280667,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004916275210379841,
      "loss": 5.4015,
      "mean_token_accuracy": 0.15935112088918685,
      "num_tokens": 21231476.0,
      "step": 9690
    },
    {
      "entropy": 5.639980030059815,
      "epoch": 0.9621873759428344,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004916179428386408,
      "loss": 5.359,
      "mean_token_accuracy": 0.16527327299118041,
      "num_tokens": 21242445.0,
      "step": 9695
    },
    {
      "entropy": 5.519500875473023,
      "epoch": 0.962683604605002,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004916083592676392,
      "loss": 5.1606,
      "mean_token_accuracy": 0.16533677130937577,
      "num_tokens": 21253179.0,
      "step": 9700
    },
    {
      "entropy": 5.5240099906921385,
      "epoch": 0.9631798332671695,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004915987703252171,
      "loss": 5.2015,
      "mean_token_accuracy": 0.1710731193423271,
      "num_tokens": 21263460.0,
      "step": 9705
    },
    {
      "entropy": 5.451681900024414,
      "epoch": 0.9636760619293371,
      "grad_norm": 0.875,
      "learning_rate": 0.0004915891760116123,
      "loss": 5.2632,
      "mean_token_accuracy": 0.1635735422372818,
      "num_tokens": 21274819.0,
      "step": 9710
    },
    {
      "entropy": 5.517275714874268,
      "epoch": 0.9641722905915046,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004915795763270626,
      "loss": 5.2254,
      "mean_token_accuracy": 0.16332022994756698,
      "num_tokens": 21286215.0,
      "step": 9715
    },
    {
      "entropy": 5.557509994506836,
      "epoch": 0.9646685192536721,
      "grad_norm": 0.953125,
      "learning_rate": 0.000491569971271806,
      "loss": 5.2211,
      "mean_token_accuracy": 0.16662717461586,
      "num_tokens": 21297033.0,
      "step": 9720
    },
    {
      "entropy": 5.490494346618652,
      "epoch": 0.9651647479158396,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004915603608460809,
      "loss": 5.2213,
      "mean_token_accuracy": 0.16428214311599731,
      "num_tokens": 21308019.0,
      "step": 9725
    },
    {
      "entropy": 5.536657381057739,
      "epoch": 0.9656609765780071,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004915507450501254,
      "loss": 5.3577,
      "mean_token_accuracy": 0.15414009913802146,
      "num_tokens": 21319313.0,
      "step": 9730
    },
    {
      "entropy": 5.569180727005005,
      "epoch": 0.9661572052401747,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000491541123884178,
      "loss": 5.2682,
      "mean_token_accuracy": 0.16304835081100463,
      "num_tokens": 21329950.0,
      "step": 9735
    },
    {
      "entropy": 5.537885951995849,
      "epoch": 0.9666534339023422,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004915314973484775,
      "loss": 5.2301,
      "mean_token_accuracy": 0.16611171811819075,
      "num_tokens": 21339061.0,
      "step": 9740
    },
    {
      "entropy": 5.530014896392823,
      "epoch": 0.9671496625645097,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004915218654432624,
      "loss": 5.3287,
      "mean_token_accuracy": 0.1598948985338211,
      "num_tokens": 21350114.0,
      "step": 9745
    },
    {
      "entropy": 5.555548477172851,
      "epoch": 0.9676458912266772,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004915122281687716,
      "loss": 5.2516,
      "mean_token_accuracy": 0.16767426282167436,
      "num_tokens": 21361219.0,
      "step": 9750
    },
    {
      "entropy": 5.485579776763916,
      "epoch": 0.9681421198888448,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004915025855252441,
      "loss": 5.2458,
      "mean_token_accuracy": 0.16266645789146422,
      "num_tokens": 21372722.0,
      "step": 9755
    },
    {
      "entropy": 5.5408666133880615,
      "epoch": 0.9686383485510123,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004914929375129188,
      "loss": 5.2766,
      "mean_token_accuracy": 0.16189596205949783,
      "num_tokens": 21383350.0,
      "step": 9760
    },
    {
      "entropy": 5.569598627090454,
      "epoch": 0.9691345772131799,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004914832841320353,
      "loss": 5.3227,
      "mean_token_accuracy": 0.1622638449072838,
      "num_tokens": 21394656.0,
      "step": 9765
    },
    {
      "entropy": 5.544922971725464,
      "epoch": 0.9696308058753473,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000491473625382833,
      "loss": 5.3138,
      "mean_token_accuracy": 0.15711605846881865,
      "num_tokens": 21406163.0,
      "step": 9770
    },
    {
      "entropy": 5.459386777877808,
      "epoch": 0.9701270345375149,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000491463961265551,
      "loss": 5.2086,
      "mean_token_accuracy": 0.16764862686395646,
      "num_tokens": 21418703.0,
      "step": 9775
    },
    {
      "entropy": 5.505564880371094,
      "epoch": 0.9706232631996824,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004914542917804292,
      "loss": 5.2119,
      "mean_token_accuracy": 0.17257481962442398,
      "num_tokens": 21428511.0,
      "step": 9780
    },
    {
      "entropy": 5.557694387435913,
      "epoch": 0.97111949186185,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004914446169277075,
      "loss": 5.2502,
      "mean_token_accuracy": 0.16785378605127335,
      "num_tokens": 21440350.0,
      "step": 9785
    },
    {
      "entropy": 5.60865159034729,
      "epoch": 0.9716157205240175,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004914349367076258,
      "loss": 5.346,
      "mean_token_accuracy": 0.1607807070016861,
      "num_tokens": 21451262.0,
      "step": 9790
    },
    {
      "entropy": 5.567680978775025,
      "epoch": 0.972111949186185,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004914252511204238,
      "loss": 5.3135,
      "mean_token_accuracy": 0.1643596589565277,
      "num_tokens": 21462587.0,
      "step": 9795
    },
    {
      "entropy": 5.5603399753570555,
      "epoch": 0.9726081778483525,
      "grad_norm": 1.0,
      "learning_rate": 0.0004914155601663422,
      "loss": 5.2482,
      "mean_token_accuracy": 0.16613210290670394,
      "num_tokens": 21473452.0,
      "step": 9800
    },
    {
      "entropy": 5.5556717872619625,
      "epoch": 0.97310440651052,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004914058638456208,
      "loss": 5.2258,
      "mean_token_accuracy": 0.16170699894428253,
      "num_tokens": 21484011.0,
      "step": 9805
    },
    {
      "entropy": 5.466827297210694,
      "epoch": 0.9736006351726876,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004913961621585003,
      "loss": 5.2195,
      "mean_token_accuracy": 0.1644126608967781,
      "num_tokens": 21495959.0,
      "step": 9810
    },
    {
      "entropy": 5.481923246383667,
      "epoch": 0.974096863834855,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004913864551052214,
      "loss": 5.2589,
      "mean_token_accuracy": 0.1722146451473236,
      "num_tokens": 21507496.0,
      "step": 9815
    },
    {
      "entropy": 5.621778869628907,
      "epoch": 0.9745930924970226,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004913767426860246,
      "loss": 5.3438,
      "mean_token_accuracy": 0.1552169665694237,
      "num_tokens": 21517532.0,
      "step": 9820
    },
    {
      "entropy": 5.49606409072876,
      "epoch": 0.9750893211591901,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004913670249011508,
      "loss": 5.1665,
      "mean_token_accuracy": 0.16770999878644943,
      "num_tokens": 21527969.0,
      "step": 9825
    },
    {
      "entropy": 5.5104217529296875,
      "epoch": 0.9755855498213577,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004913573017508411,
      "loss": 5.2711,
      "mean_token_accuracy": 0.15946028381586075,
      "num_tokens": 21538243.0,
      "step": 9830
    },
    {
      "entropy": 5.563217544555664,
      "epoch": 0.9760817784835252,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004913475732353365,
      "loss": 5.2803,
      "mean_token_accuracy": 0.16343535631895065,
      "num_tokens": 21548702.0,
      "step": 9835
    },
    {
      "entropy": 5.486120510101318,
      "epoch": 0.9765780071456928,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004913378393548784,
      "loss": 5.2372,
      "mean_token_accuracy": 0.16483518332242966,
      "num_tokens": 21558583.0,
      "step": 9840
    },
    {
      "entropy": 5.504124307632447,
      "epoch": 0.9770742358078602,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004913281001097078,
      "loss": 5.2229,
      "mean_token_accuracy": 0.16806641072034836,
      "num_tokens": 21569845.0,
      "step": 9845
    },
    {
      "entropy": 5.567736673355102,
      "epoch": 0.9775704644700278,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004913183555000667,
      "loss": 5.3681,
      "mean_token_accuracy": 0.15614749044179915,
      "num_tokens": 21580540.0,
      "step": 9850
    },
    {
      "entropy": 5.568388223648071,
      "epoch": 0.9780666931321953,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004913086055261963,
      "loss": 5.2303,
      "mean_token_accuracy": 0.1666613608598709,
      "num_tokens": 21590982.0,
      "step": 9855
    },
    {
      "entropy": 5.524022960662842,
      "epoch": 0.9785629217943629,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004912988501883387,
      "loss": 5.2565,
      "mean_token_accuracy": 0.16447547525167466,
      "num_tokens": 21602817.0,
      "step": 9860
    },
    {
      "entropy": 5.5490241050720215,
      "epoch": 0.9790591504565304,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004912890894867357,
      "loss": 5.3001,
      "mean_token_accuracy": 0.16047051697969436,
      "num_tokens": 21613422.0,
      "step": 9865
    },
    {
      "entropy": 5.536137580871582,
      "epoch": 0.9795553791186979,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004912793234216293,
      "loss": 5.2961,
      "mean_token_accuracy": 0.1544037289917469,
      "num_tokens": 21624241.0,
      "step": 9870
    },
    {
      "entropy": 5.485258722305298,
      "epoch": 0.9800516077808654,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004912695519932618,
      "loss": 5.2854,
      "mean_token_accuracy": 0.164899580180645,
      "num_tokens": 21636243.0,
      "step": 9875
    },
    {
      "entropy": 5.5323365211486815,
      "epoch": 0.9805478364430329,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004912597752018753,
      "loss": 5.3106,
      "mean_token_accuracy": 0.15859518200159073,
      "num_tokens": 21647918.0,
      "step": 9880
    },
    {
      "entropy": 5.5906257152557375,
      "epoch": 0.9810440651052005,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004912499930477126,
      "loss": 5.2838,
      "mean_token_accuracy": 0.16075822114944457,
      "num_tokens": 21659236.0,
      "step": 9885
    },
    {
      "entropy": 5.5201832294464115,
      "epoch": 0.981540293767368,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004912402055310158,
      "loss": 5.2809,
      "mean_token_accuracy": 0.1637316793203354,
      "num_tokens": 21669647.0,
      "step": 9890
    },
    {
      "entropy": 5.509677886962891,
      "epoch": 0.9820365224295355,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004912304126520279,
      "loss": 5.3217,
      "mean_token_accuracy": 0.15920768976211547,
      "num_tokens": 21680369.0,
      "step": 9895
    },
    {
      "entropy": 5.593974637985229,
      "epoch": 0.982532751091703,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004912206144109917,
      "loss": 5.3321,
      "mean_token_accuracy": 0.15918789356946944,
      "num_tokens": 21690942.0,
      "step": 9900
    },
    {
      "entropy": 5.552346706390381,
      "epoch": 0.9830289797538706,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004912108108081503,
      "loss": 5.2555,
      "mean_token_accuracy": 0.16583765149116517,
      "num_tokens": 21701945.0,
      "step": 9905
    },
    {
      "entropy": 5.577497863769532,
      "epoch": 0.9835252084160381,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004912010018437465,
      "loss": 5.284,
      "mean_token_accuracy": 0.1665597915649414,
      "num_tokens": 21712246.0,
      "step": 9910
    },
    {
      "entropy": 5.6315020561218265,
      "epoch": 0.9840214370782057,
      "grad_norm": 0.890625,
      "learning_rate": 0.000491191187518024,
      "loss": 5.2855,
      "mean_token_accuracy": 0.16163774877786635,
      "num_tokens": 21721660.0,
      "step": 9915
    },
    {
      "entropy": 5.506907224655151,
      "epoch": 0.9845176657403731,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004911813678312257,
      "loss": 5.17,
      "mean_token_accuracy": 0.1689017653465271,
      "num_tokens": 21732157.0,
      "step": 9920
    },
    {
      "entropy": 5.512174034118653,
      "epoch": 0.9850138944025407,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004911715427835953,
      "loss": 5.3077,
      "mean_token_accuracy": 0.15705085918307304,
      "num_tokens": 21741727.0,
      "step": 9925
    },
    {
      "entropy": 5.540558862686157,
      "epoch": 0.9855101230647082,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004911617123753765,
      "loss": 5.2454,
      "mean_token_accuracy": 0.1678730845451355,
      "num_tokens": 21752596.0,
      "step": 9930
    },
    {
      "entropy": 5.4900116443634035,
      "epoch": 0.9860063517268758,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004911518766068131,
      "loss": 5.1613,
      "mean_token_accuracy": 0.16437255591154099,
      "num_tokens": 21763978.0,
      "step": 9935
    },
    {
      "entropy": 5.518296909332276,
      "epoch": 0.9865025803890433,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004911420354781487,
      "loss": 5.3046,
      "mean_token_accuracy": 0.15439167469739914,
      "num_tokens": 21774961.0,
      "step": 9940
    },
    {
      "entropy": 5.527509880065918,
      "epoch": 0.9869988090512108,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004911321889896278,
      "loss": 5.2118,
      "mean_token_accuracy": 0.17294728606939316,
      "num_tokens": 21785387.0,
      "step": 9945
    },
    {
      "entropy": 5.514861297607422,
      "epoch": 0.9874950377133783,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004911223371414943,
      "loss": 5.2672,
      "mean_token_accuracy": 0.16925038099288942,
      "num_tokens": 21795982.0,
      "step": 9950
    },
    {
      "entropy": 5.586288785934448,
      "epoch": 0.9879912663755459,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004911124799339924,
      "loss": 5.3018,
      "mean_token_accuracy": 0.15709125995635986,
      "num_tokens": 21806768.0,
      "step": 9955
    },
    {
      "entropy": 5.5430042266845705,
      "epoch": 0.9884874950377134,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004911026173673667,
      "loss": 5.2655,
      "mean_token_accuracy": 0.16700110733509063,
      "num_tokens": 21818563.0,
      "step": 9960
    },
    {
      "entropy": 5.57335844039917,
      "epoch": 0.9889837236998809,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004910927494418619,
      "loss": 5.3358,
      "mean_token_accuracy": 0.16360723674297334,
      "num_tokens": 21830253.0,
      "step": 9965
    },
    {
      "entropy": 5.5798474788665775,
      "epoch": 0.9894799523620484,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004910828761577224,
      "loss": 5.3094,
      "mean_token_accuracy": 0.1650498926639557,
      "num_tokens": 21842353.0,
      "step": 9970
    },
    {
      "entropy": 5.530132532119751,
      "epoch": 0.9899761810242159,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004910729975151933,
      "loss": 5.2442,
      "mean_token_accuracy": 0.1677899479866028,
      "num_tokens": 21852066.0,
      "step": 9975
    },
    {
      "entropy": 5.504552602767944,
      "epoch": 0.9904724096863835,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004910631135145194,
      "loss": 5.2664,
      "mean_token_accuracy": 0.16598305702209473,
      "num_tokens": 21862754.0,
      "step": 9980
    },
    {
      "entropy": 5.441079378128052,
      "epoch": 0.990968638348551,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004910532241559458,
      "loss": 5.1758,
      "mean_token_accuracy": 0.17741164416074753,
      "num_tokens": 21873531.0,
      "step": 9985
    },
    {
      "entropy": 5.444345712661743,
      "epoch": 0.9914648670107186,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004910433294397179,
      "loss": 5.1451,
      "mean_token_accuracy": 0.16984826773405076,
      "num_tokens": 21884339.0,
      "step": 9990
    },
    {
      "entropy": 5.520063018798828,
      "epoch": 0.991961095672886,
      "grad_norm": 0.921875,
      "learning_rate": 0.000491033429366081,
      "loss": 5.2445,
      "mean_token_accuracy": 0.16473416090011597,
      "num_tokens": 21895489.0,
      "step": 9995
    },
    {
      "entropy": 5.527721643447876,
      "epoch": 0.9924573243350536,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004910235239352804,
      "loss": 5.2228,
      "mean_token_accuracy": 0.1649106651544571,
      "num_tokens": 21906832.0,
      "step": 10000
    },
    {
      "entropy": 5.516564750671387,
      "epoch": 0.9929535529972211,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000491013613147562,
      "loss": 5.2757,
      "mean_token_accuracy": 0.16007844358682632,
      "num_tokens": 21916979.0,
      "step": 10005
    },
    {
      "entropy": 5.515347290039062,
      "epoch": 0.9934497816593887,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004910036970031715,
      "loss": 5.1807,
      "mean_token_accuracy": 0.1732892245054245,
      "num_tokens": 21927352.0,
      "step": 10010
    },
    {
      "entropy": 5.514059352874756,
      "epoch": 0.9939460103215562,
      "grad_norm": 0.875,
      "learning_rate": 0.0004909937755023547,
      "loss": 5.3066,
      "mean_token_accuracy": 0.1569930851459503,
      "num_tokens": 21937765.0,
      "step": 10015
    },
    {
      "entropy": 5.46819486618042,
      "epoch": 0.9944422389837237,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004909838486453577,
      "loss": 5.1887,
      "mean_token_accuracy": 0.16520598977804185,
      "num_tokens": 21948659.0,
      "step": 10020
    },
    {
      "entropy": 5.517417621612549,
      "epoch": 0.9949384676458912,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004909739164324267,
      "loss": 5.3063,
      "mean_token_accuracy": 0.15833762884140015,
      "num_tokens": 21959356.0,
      "step": 10025
    },
    {
      "entropy": 5.654903221130371,
      "epoch": 0.9954346963080588,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004909639788638079,
      "loss": 5.4189,
      "mean_token_accuracy": 0.15256367325782777,
      "num_tokens": 21971321.0,
      "step": 10030
    },
    {
      "entropy": 5.534298658370972,
      "epoch": 0.9959309249702263,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004909540359397477,
      "loss": 5.2701,
      "mean_token_accuracy": 0.16025426238775253,
      "num_tokens": 21981252.0,
      "step": 10035
    },
    {
      "entropy": 5.559725046157837,
      "epoch": 0.9964271536323938,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004909440876604929,
      "loss": 5.2782,
      "mean_token_accuracy": 0.15571915805339814,
      "num_tokens": 21991820.0,
      "step": 10040
    },
    {
      "entropy": 5.531323003768921,
      "epoch": 0.9969233822945613,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00049093413402629,
      "loss": 5.2442,
      "mean_token_accuracy": 0.16417673975229263,
      "num_tokens": 22003619.0,
      "step": 10045
    },
    {
      "entropy": 5.542947816848755,
      "epoch": 0.9974196109567288,
      "grad_norm": 1.0,
      "learning_rate": 0.0004909241750373858,
      "loss": 5.2973,
      "mean_token_accuracy": 0.15834089517593383,
      "num_tokens": 22016116.0,
      "step": 10050
    },
    {
      "entropy": 5.580583333969116,
      "epoch": 0.9979158396188964,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004909142106940275,
      "loss": 5.2283,
      "mean_token_accuracy": 0.1644928440451622,
      "num_tokens": 22026115.0,
      "step": 10055
    },
    {
      "entropy": 5.565815401077271,
      "epoch": 0.9984120682810639,
      "grad_norm": 0.9375,
      "learning_rate": 0.000490904240996462,
      "loss": 5.2981,
      "mean_token_accuracy": 0.16727686673402786,
      "num_tokens": 22038889.0,
      "step": 10060
    },
    {
      "entropy": 5.524577999114991,
      "epoch": 0.9989082969432315,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004908942659449364,
      "loss": 5.2504,
      "mean_token_accuracy": 0.1659238651394844,
      "num_tokens": 22050773.0,
      "step": 10065
    },
    {
      "entropy": 5.491831970214844,
      "epoch": 0.9994045256053989,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004908842855396982,
      "loss": 5.2598,
      "mean_token_accuracy": 0.15760456025600433,
      "num_tokens": 22062869.0,
      "step": 10070
    },
    {
      "entropy": 5.530044889450073,
      "epoch": 0.9999007542675665,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004908742997809951,
      "loss": 5.2714,
      "mean_token_accuracy": 0.16301410943269729,
      "num_tokens": 22073898.0,
      "step": 10075
    },
    {
      "entropy": 5.574612617492676,
      "epoch": 1.0003969829297341,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004908643086690744,
      "loss": 5.1844,
      "mean_token_accuracy": 0.16193026006221772,
      "num_tokens": 22084383.0,
      "step": 10080
    },
    {
      "entropy": 5.51653642654419,
      "epoch": 1.0008932115919016,
      "grad_norm": 0.921875,
      "learning_rate": 0.000490854312204184,
      "loss": 5.2201,
      "mean_token_accuracy": 0.16356946527957916,
      "num_tokens": 22094990.0,
      "step": 10085
    },
    {
      "entropy": 5.531592988967896,
      "epoch": 1.001389440254069,
      "grad_norm": 0.875,
      "learning_rate": 0.0004908443103865719,
      "loss": 5.2525,
      "mean_token_accuracy": 0.16223589330911636,
      "num_tokens": 22105259.0,
      "step": 10090
    },
    {
      "entropy": 5.529381132125854,
      "epoch": 1.0018856689162365,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004908343032164859,
      "loss": 5.218,
      "mean_token_accuracy": 0.16315143704414367,
      "num_tokens": 22116348.0,
      "step": 10095
    },
    {
      "entropy": 5.544572830200195,
      "epoch": 1.0023818975784042,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004908242906941743,
      "loss": 5.2082,
      "mean_token_accuracy": 0.16781669706106186,
      "num_tokens": 22127311.0,
      "step": 10100
    },
    {
      "entropy": 5.584514999389649,
      "epoch": 1.0028781262405717,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004908142728198853,
      "loss": 5.2308,
      "mean_token_accuracy": 0.16152353584766388,
      "num_tokens": 22137929.0,
      "step": 10105
    },
    {
      "entropy": 5.626436710357666,
      "epoch": 1.0033743549027392,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004908042495938673,
      "loss": 5.3409,
      "mean_token_accuracy": 0.16043704748153687,
      "num_tokens": 22148202.0,
      "step": 10110
    },
    {
      "entropy": 5.509261560440064,
      "epoch": 1.0038705835649067,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004907942210163692,
      "loss": 5.139,
      "mean_token_accuracy": 0.16677588373422622,
      "num_tokens": 22158955.0,
      "step": 10115
    },
    {
      "entropy": 5.517710399627686,
      "epoch": 1.0043668122270741,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004907841870876392,
      "loss": 5.1703,
      "mean_token_accuracy": 0.16501835733652115,
      "num_tokens": 22169569.0,
      "step": 10120
    },
    {
      "entropy": 5.489158821105957,
      "epoch": 1.0048630408892418,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004907741478079263,
      "loss": 5.1477,
      "mean_token_accuracy": 0.16425950676202775,
      "num_tokens": 22180171.0,
      "step": 10125
    },
    {
      "entropy": 5.505987215042114,
      "epoch": 1.0053592695514093,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004907641031774796,
      "loss": 5.2712,
      "mean_token_accuracy": 0.16013553142547607,
      "num_tokens": 22190531.0,
      "step": 10130
    },
    {
      "entropy": 5.506381320953369,
      "epoch": 1.0058554982135768,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000490754053196548,
      "loss": 5.1612,
      "mean_token_accuracy": 0.16621319204568863,
      "num_tokens": 22201495.0,
      "step": 10135
    },
    {
      "entropy": 5.585003519058228,
      "epoch": 1.0063517268757443,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004907439978653809,
      "loss": 5.2439,
      "mean_token_accuracy": 0.15559897422790528,
      "num_tokens": 22212941.0,
      "step": 10140
    },
    {
      "entropy": 5.596948575973511,
      "epoch": 1.006847955537912,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004907339371842275,
      "loss": 5.2728,
      "mean_token_accuracy": 0.1628827027976513,
      "num_tokens": 22225212.0,
      "step": 10145
    },
    {
      "entropy": 5.503411531448364,
      "epoch": 1.0073441842000794,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004907238711533373,
      "loss": 5.2133,
      "mean_token_accuracy": 0.1630281016230583,
      "num_tokens": 22236370.0,
      "step": 10150
    },
    {
      "entropy": 5.5146815299987795,
      "epoch": 1.007840412862247,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004907137997729599,
      "loss": 5.1852,
      "mean_token_accuracy": 0.16718839406967162,
      "num_tokens": 22247641.0,
      "step": 10155
    },
    {
      "entropy": 5.473132514953614,
      "epoch": 1.0083366415244144,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000490703723043345,
      "loss": 5.2471,
      "mean_token_accuracy": 0.166613307595253,
      "num_tokens": 22258599.0,
      "step": 10160
    },
    {
      "entropy": 5.531865882873535,
      "epoch": 1.008832870186582,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004906936409647428,
      "loss": 5.1721,
      "mean_token_accuracy": 0.1649941012263298,
      "num_tokens": 22270913.0,
      "step": 10165
    },
    {
      "entropy": 5.511922311782837,
      "epoch": 1.0093290988487496,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000490683553537403,
      "loss": 5.0885,
      "mean_token_accuracy": 0.16907574087381363,
      "num_tokens": 22281053.0,
      "step": 10170
    },
    {
      "entropy": 5.4852986335754395,
      "epoch": 1.009825327510917,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004906734607615758,
      "loss": 5.0978,
      "mean_token_accuracy": 0.1731666073203087,
      "num_tokens": 22292570.0,
      "step": 10175
    },
    {
      "entropy": 5.496685314178467,
      "epoch": 1.0103215561730845,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004906633626375114,
      "loss": 5.1819,
      "mean_token_accuracy": 0.16467416435480117,
      "num_tokens": 22303548.0,
      "step": 10180
    },
    {
      "entropy": 5.523727798461914,
      "epoch": 1.010817784835252,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004906532591654604,
      "loss": 5.2105,
      "mean_token_accuracy": 0.16738889664411544,
      "num_tokens": 22314133.0,
      "step": 10185
    },
    {
      "entropy": 5.545677757263183,
      "epoch": 1.0113140134974197,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004906431503456733,
      "loss": 5.1641,
      "mean_token_accuracy": 0.1664816200733185,
      "num_tokens": 22326431.0,
      "step": 10190
    },
    {
      "entropy": 5.537136840820312,
      "epoch": 1.0118102421595871,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004906330361784007,
      "loss": 5.218,
      "mean_token_accuracy": 0.16480584293603898,
      "num_tokens": 22338507.0,
      "step": 10195
    },
    {
      "entropy": 5.5465387344360355,
      "epoch": 1.0123064708217546,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004906229166638936,
      "loss": 5.1878,
      "mean_token_accuracy": 0.16493188440799714,
      "num_tokens": 22348863.0,
      "step": 10200
    },
    {
      "entropy": 5.474705839157105,
      "epoch": 1.012802699483922,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004906127918024026,
      "loss": 5.139,
      "mean_token_accuracy": 0.1639999732375145,
      "num_tokens": 22359000.0,
      "step": 10205
    },
    {
      "entropy": 5.56318154335022,
      "epoch": 1.0132989281460898,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004906026615941791,
      "loss": 5.2609,
      "mean_token_accuracy": 0.16299256980419158,
      "num_tokens": 22369539.0,
      "step": 10210
    },
    {
      "entropy": 5.552541208267212,
      "epoch": 1.0137951568082573,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004905925260394741,
      "loss": 5.2348,
      "mean_token_accuracy": 0.163515704870224,
      "num_tokens": 22380700.0,
      "step": 10215
    },
    {
      "entropy": 5.525241994857788,
      "epoch": 1.0142913854704247,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000490582385138539,
      "loss": 5.211,
      "mean_token_accuracy": 0.1586480900645256,
      "num_tokens": 22392151.0,
      "step": 10220
    },
    {
      "entropy": 5.4952363014221195,
      "epoch": 1.0147876141325922,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004905722388916254,
      "loss": 5.188,
      "mean_token_accuracy": 0.17093088775873183,
      "num_tokens": 22401735.0,
      "step": 10225
    },
    {
      "entropy": 5.524148988723755,
      "epoch": 1.01528384279476,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004905620872989846,
      "loss": 5.2472,
      "mean_token_accuracy": 0.1665743976831436,
      "num_tokens": 22412549.0,
      "step": 10230
    },
    {
      "entropy": 5.612970018386841,
      "epoch": 1.0157800714569274,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004905519303608687,
      "loss": 5.2171,
      "mean_token_accuracy": 0.16105247288942337,
      "num_tokens": 22424728.0,
      "step": 10235
    },
    {
      "entropy": 5.472014379501343,
      "epoch": 1.0162763001190949,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004905417680775294,
      "loss": 5.1082,
      "mean_token_accuracy": 0.16330609768629073,
      "num_tokens": 22434197.0,
      "step": 10240
    },
    {
      "entropy": 5.578816938400268,
      "epoch": 1.0167725287812623,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004905316004492186,
      "loss": 5.2974,
      "mean_token_accuracy": 0.15572961866855622,
      "num_tokens": 22444671.0,
      "step": 10245
    },
    {
      "entropy": 5.495514726638794,
      "epoch": 1.01726875744343,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004905214274761885,
      "loss": 5.1684,
      "mean_token_accuracy": 0.1653987407684326,
      "num_tokens": 22456636.0,
      "step": 10250
    },
    {
      "entropy": 5.4824018478393555,
      "epoch": 1.0177649861055975,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004905112491586915,
      "loss": 5.1076,
      "mean_token_accuracy": 0.17306455075740815,
      "num_tokens": 22467011.0,
      "step": 10255
    },
    {
      "entropy": 5.50503191947937,
      "epoch": 1.018261214767765,
      "grad_norm": 0.96875,
      "learning_rate": 0.00049050106549698,
      "loss": 5.2013,
      "mean_token_accuracy": 0.16318316012620926,
      "num_tokens": 22477630.0,
      "step": 10260
    },
    {
      "entropy": 5.486586093902588,
      "epoch": 1.0187574434299325,
      "grad_norm": 1.0,
      "learning_rate": 0.0004904908764913063,
      "loss": 5.1528,
      "mean_token_accuracy": 0.17257197499275206,
      "num_tokens": 22488132.0,
      "step": 10265
    },
    {
      "entropy": 5.468431901931763,
      "epoch": 1.0192536720921,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004904806821419233,
      "loss": 5.2124,
      "mean_token_accuracy": 0.16477933526039124,
      "num_tokens": 22499430.0,
      "step": 10270
    },
    {
      "entropy": 5.486281061172486,
      "epoch": 1.0197499007542676,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004904704824490835,
      "loss": 5.1387,
      "mean_token_accuracy": 0.16794027537107467,
      "num_tokens": 22509728.0,
      "step": 10275
    },
    {
      "entropy": 5.530235671997071,
      "epoch": 1.020246129416435,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004904602774130402,
      "loss": 5.2485,
      "mean_token_accuracy": 0.15583620667457582,
      "num_tokens": 22520955.0,
      "step": 10280
    },
    {
      "entropy": 5.578283500671387,
      "epoch": 1.0207423580786026,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004904500670340463,
      "loss": 5.2563,
      "mean_token_accuracy": 0.15621759369969368,
      "num_tokens": 22530440.0,
      "step": 10285
    },
    {
      "entropy": 5.51908540725708,
      "epoch": 1.02123858674077,
      "grad_norm": 0.859375,
      "learning_rate": 0.000490439851312355,
      "loss": 5.2293,
      "mean_token_accuracy": 0.17005994617938996,
      "num_tokens": 22541539.0,
      "step": 10290
    },
    {
      "entropy": 5.478838062286377,
      "epoch": 1.0217348154029378,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004904296302482196,
      "loss": 5.2226,
      "mean_token_accuracy": 0.164718359708786,
      "num_tokens": 22555717.0,
      "step": 10295
    },
    {
      "entropy": 5.59019660949707,
      "epoch": 1.0222310440651052,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004904194038418935,
      "loss": 5.3002,
      "mean_token_accuracy": 0.15941984802484513,
      "num_tokens": 22566938.0,
      "step": 10300
    },
    {
      "entropy": 5.628157997131348,
      "epoch": 1.0227272727272727,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004904091720936305,
      "loss": 5.1681,
      "mean_token_accuracy": 0.16496185809373856,
      "num_tokens": 22578725.0,
      "step": 10305
    },
    {
      "entropy": 5.531756830215454,
      "epoch": 1.0232235013894402,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004903989350036841,
      "loss": 5.0983,
      "mean_token_accuracy": 0.17594655752182006,
      "num_tokens": 22588476.0,
      "step": 10310
    },
    {
      "entropy": 5.53554630279541,
      "epoch": 1.0237197300516079,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004903886925723083,
      "loss": 5.3216,
      "mean_token_accuracy": 0.1539576232433319,
      "num_tokens": 22600294.0,
      "step": 10315
    },
    {
      "entropy": 5.569748830795288,
      "epoch": 1.0242159587137754,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000490378444799757,
      "loss": 5.198,
      "mean_token_accuracy": 0.16748596355319023,
      "num_tokens": 22610578.0,
      "step": 10320
    },
    {
      "entropy": 5.591857004165649,
      "epoch": 1.0247121873759428,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004903681916862844,
      "loss": 5.2055,
      "mean_token_accuracy": 0.16512291729450226,
      "num_tokens": 22620988.0,
      "step": 10325
    },
    {
      "entropy": 5.520949745178223,
      "epoch": 1.0252084160381103,
      "grad_norm": 0.875,
      "learning_rate": 0.0004903579332321447,
      "loss": 5.1952,
      "mean_token_accuracy": 0.1655699610710144,
      "num_tokens": 22632198.0,
      "step": 10330
    },
    {
      "entropy": 5.524732303619385,
      "epoch": 1.0257046447002778,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004903476694375924,
      "loss": 5.201,
      "mean_token_accuracy": 0.17436613887548447,
      "num_tokens": 22642543.0,
      "step": 10335
    },
    {
      "entropy": 5.542492485046386,
      "epoch": 1.0262008733624455,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004903374003028819,
      "loss": 5.2051,
      "mean_token_accuracy": 0.16694827228784562,
      "num_tokens": 22655453.0,
      "step": 10340
    },
    {
      "entropy": 5.478806352615356,
      "epoch": 1.026697102024613,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004903271258282678,
      "loss": 5.1442,
      "mean_token_accuracy": 0.17118590772151948,
      "num_tokens": 22667342.0,
      "step": 10345
    },
    {
      "entropy": 5.581794404983521,
      "epoch": 1.0271933306867804,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004903168460140049,
      "loss": 5.2131,
      "mean_token_accuracy": 0.17006220370531083,
      "num_tokens": 22676890.0,
      "step": 10350
    },
    {
      "entropy": 5.539871406555176,
      "epoch": 1.027689559348948,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004903065608603484,
      "loss": 5.2111,
      "mean_token_accuracy": 0.16357226967811583,
      "num_tokens": 22687809.0,
      "step": 10355
    },
    {
      "entropy": 5.488600587844848,
      "epoch": 1.0281857880111156,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000490296270367553,
      "loss": 5.224,
      "mean_token_accuracy": 0.16250108629465104,
      "num_tokens": 22697938.0,
      "step": 10360
    },
    {
      "entropy": 5.61170334815979,
      "epoch": 1.028682016673283,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004902859745358741,
      "loss": 5.2199,
      "mean_token_accuracy": 0.17221857160329818,
      "num_tokens": 22708303.0,
      "step": 10365
    },
    {
      "entropy": 5.50864953994751,
      "epoch": 1.0291782453354505,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004902756733655668,
      "loss": 5.1535,
      "mean_token_accuracy": 0.1687371775507927,
      "num_tokens": 22717987.0,
      "step": 10370
    },
    {
      "entropy": 5.512455415725708,
      "epoch": 1.029674473997618,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004902653668568868,
      "loss": 5.2313,
      "mean_token_accuracy": 0.16371186822652817,
      "num_tokens": 22729672.0,
      "step": 10375
    },
    {
      "entropy": 5.576651382446289,
      "epoch": 1.0301707026597857,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004902550550100896,
      "loss": 5.1842,
      "mean_token_accuracy": 0.17124547809362411,
      "num_tokens": 22741296.0,
      "step": 10380
    },
    {
      "entropy": 5.529744672775268,
      "epoch": 1.0306669313219532,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004902447378254307,
      "loss": 5.1767,
      "mean_token_accuracy": 0.1712263137102127,
      "num_tokens": 22752337.0,
      "step": 10385
    },
    {
      "entropy": 5.553467845916748,
      "epoch": 1.0311631599841207,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004902344153031663,
      "loss": 5.2157,
      "mean_token_accuracy": 0.16358236074447632,
      "num_tokens": 22762681.0,
      "step": 10390
    },
    {
      "entropy": 5.600700044631958,
      "epoch": 1.0316593886462881,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000490224087443552,
      "loss": 5.2636,
      "mean_token_accuracy": 0.15973918735980988,
      "num_tokens": 22774275.0,
      "step": 10395
    },
    {
      "entropy": 5.502658748626709,
      "epoch": 1.0321556173084558,
      "grad_norm": 1.0,
      "learning_rate": 0.0004902137542468442,
      "loss": 5.1395,
      "mean_token_accuracy": 0.17113078981637955,
      "num_tokens": 22785944.0,
      "step": 10400
    },
    {
      "entropy": 5.451526737213134,
      "epoch": 1.0326518459706233,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004902034157132991,
      "loss": 5.0633,
      "mean_token_accuracy": 0.16619019359350204,
      "num_tokens": 22796543.0,
      "step": 10405
    },
    {
      "entropy": 5.454790830612183,
      "epoch": 1.0331480746327908,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004901930718431729,
      "loss": 5.1876,
      "mean_token_accuracy": 0.16527335494756698,
      "num_tokens": 22807957.0,
      "step": 10410
    },
    {
      "entropy": 5.563516902923584,
      "epoch": 1.0336443032949583,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004901827226367223,
      "loss": 5.2828,
      "mean_token_accuracy": 0.16433306634426117,
      "num_tokens": 22818903.0,
      "step": 10415
    },
    {
      "entropy": 5.579446649551391,
      "epoch": 1.0341405319571257,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004901723680942038,
      "loss": 5.3108,
      "mean_token_accuracy": 0.16274386048316955,
      "num_tokens": 22829286.0,
      "step": 10420
    },
    {
      "entropy": 5.599664211273193,
      "epoch": 1.0346367606192934,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004901620082158742,
      "loss": 5.2202,
      "mean_token_accuracy": 0.16135814189910888,
      "num_tokens": 22839860.0,
      "step": 10425
    },
    {
      "entropy": 5.625802993774414,
      "epoch": 1.035132989281461,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004901516430019905,
      "loss": 5.2771,
      "mean_token_accuracy": 0.15363107174634932,
      "num_tokens": 22852238.0,
      "step": 10430
    },
    {
      "entropy": 5.5464558601379395,
      "epoch": 1.0356292179436284,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004901412724528097,
      "loss": 5.2357,
      "mean_token_accuracy": 0.1671655386686325,
      "num_tokens": 22863013.0,
      "step": 10435
    },
    {
      "entropy": 5.477605962753296,
      "epoch": 1.0361254466057959,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004901308965685889,
      "loss": 5.0898,
      "mean_token_accuracy": 0.166864375770092,
      "num_tokens": 22874233.0,
      "step": 10440
    },
    {
      "entropy": 5.533360528945923,
      "epoch": 1.0366216752679636,
      "grad_norm": 0.875,
      "learning_rate": 0.0004901205153495853,
      "loss": 5.273,
      "mean_token_accuracy": 0.16261522918939592,
      "num_tokens": 22885313.0,
      "step": 10445
    },
    {
      "entropy": 5.524601840972901,
      "epoch": 1.037117903930131,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004901101287960566,
      "loss": 5.1543,
      "mean_token_accuracy": 0.16824844479560852,
      "num_tokens": 22895365.0,
      "step": 10450
    },
    {
      "entropy": 5.499677896499634,
      "epoch": 1.0376141325922985,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004900997369082602,
      "loss": 5.1551,
      "mean_token_accuracy": 0.16431809812784196,
      "num_tokens": 22906392.0,
      "step": 10455
    },
    {
      "entropy": 5.454514503479004,
      "epoch": 1.038110361254466,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004900893396864539,
      "loss": 5.1099,
      "mean_token_accuracy": 0.17248469442129136,
      "num_tokens": 22916760.0,
      "step": 10460
    },
    {
      "entropy": 5.541943502426148,
      "epoch": 1.0386065899166337,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004900789371308953,
      "loss": 5.1727,
      "mean_token_accuracy": 0.16550545692443847,
      "num_tokens": 22928070.0,
      "step": 10465
    },
    {
      "entropy": 5.448971796035766,
      "epoch": 1.0391028185788012,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004900685292418425,
      "loss": 5.1535,
      "mean_token_accuracy": 0.1698997437953949,
      "num_tokens": 22938815.0,
      "step": 10470
    },
    {
      "entropy": 5.519144487380982,
      "epoch": 1.0395990472409686,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004900581160195537,
      "loss": 5.228,
      "mean_token_accuracy": 0.15901804566383362,
      "num_tokens": 22949806.0,
      "step": 10475
    },
    {
      "entropy": 5.454152297973633,
      "epoch": 1.040095275903136,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000490047697464287,
      "loss": 5.0665,
      "mean_token_accuracy": 0.16806019246578216,
      "num_tokens": 22960880.0,
      "step": 10480
    },
    {
      "entropy": 5.5343646049499515,
      "epoch": 1.0405915045653038,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004900372735763009,
      "loss": 5.2213,
      "mean_token_accuracy": 0.16154549568891524,
      "num_tokens": 22972828.0,
      "step": 10485
    },
    {
      "entropy": 5.451268148422241,
      "epoch": 1.0410877332274713,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004900268443558536,
      "loss": 5.1264,
      "mean_token_accuracy": 0.1684920996427536,
      "num_tokens": 22981869.0,
      "step": 10490
    },
    {
      "entropy": 5.549597644805909,
      "epoch": 1.0415839618896388,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000490016409803204,
      "loss": 5.2251,
      "mean_token_accuracy": 0.15966662913560867,
      "num_tokens": 22993111.0,
      "step": 10495
    },
    {
      "entropy": 5.527101325988769,
      "epoch": 1.0420801905518062,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004900059699186108,
      "loss": 5.208,
      "mean_token_accuracy": 0.1624673679471016,
      "num_tokens": 23003115.0,
      "step": 10500
    },
    {
      "entropy": 5.56898045539856,
      "epoch": 1.0425764192139737,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004899955247023328,
      "loss": 5.2478,
      "mean_token_accuracy": 0.16275709122419357,
      "num_tokens": 23012587.0,
      "step": 10505
    },
    {
      "entropy": 5.5545085906982425,
      "epoch": 1.0430726478761414,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000489985074154629,
      "loss": 5.2612,
      "mean_token_accuracy": 0.16268475502729415,
      "num_tokens": 23024455.0,
      "step": 10510
    },
    {
      "entropy": 5.506721019744873,
      "epoch": 1.0435688765383089,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004899746182757587,
      "loss": 5.1938,
      "mean_token_accuracy": 0.16569384187459946,
      "num_tokens": 23034430.0,
      "step": 10515
    },
    {
      "entropy": 5.547843074798584,
      "epoch": 1.0440651052004764,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004899641570659811,
      "loss": 5.2456,
      "mean_token_accuracy": 0.16971434727311135,
      "num_tokens": 23044811.0,
      "step": 10520
    },
    {
      "entropy": 5.573384952545166,
      "epoch": 1.0445613338626438,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004899536905255555,
      "loss": 5.2911,
      "mean_token_accuracy": 0.1627610817551613,
      "num_tokens": 23056179.0,
      "step": 10525
    },
    {
      "entropy": 5.552228784561157,
      "epoch": 1.0450575625248115,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004899432186547416,
      "loss": 5.2078,
      "mean_token_accuracy": 0.1670810952782631,
      "num_tokens": 23068331.0,
      "step": 10530
    },
    {
      "entropy": 5.5211748600006105,
      "epoch": 1.045553791186979,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004899327414537991,
      "loss": 5.1866,
      "mean_token_accuracy": 0.16346872746944427,
      "num_tokens": 23079213.0,
      "step": 10535
    },
    {
      "entropy": 5.63276047706604,
      "epoch": 1.0460500198491465,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004899222589229877,
      "loss": 5.2932,
      "mean_token_accuracy": 0.1583160325884819,
      "num_tokens": 23089860.0,
      "step": 10540
    },
    {
      "entropy": 5.4699718952178955,
      "epoch": 1.046546248511314,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004899117710625674,
      "loss": 5.1626,
      "mean_token_accuracy": 0.17572366297245026,
      "num_tokens": 23099431.0,
      "step": 10545
    },
    {
      "entropy": 5.507458305358886,
      "epoch": 1.0470424771734816,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004899012778727983,
      "loss": 5.228,
      "mean_token_accuracy": 0.1616830661892891,
      "num_tokens": 23110947.0,
      "step": 10550
    },
    {
      "entropy": 5.591244840621949,
      "epoch": 1.0475387058356491,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004898907793539406,
      "loss": 5.3145,
      "mean_token_accuracy": 0.16271693259477615,
      "num_tokens": 23122782.0,
      "step": 10555
    },
    {
      "entropy": 5.6101878643035885,
      "epoch": 1.0480349344978166,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004898802755062546,
      "loss": 5.2243,
      "mean_token_accuracy": 0.164697103202343,
      "num_tokens": 23134254.0,
      "step": 10560
    },
    {
      "entropy": 5.476492500305175,
      "epoch": 1.048531163159984,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004898697663300007,
      "loss": 5.0673,
      "mean_token_accuracy": 0.17472380995750428,
      "num_tokens": 23143851.0,
      "step": 10565
    },
    {
      "entropy": 5.486402606964111,
      "epoch": 1.0490273918221515,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004898592518254398,
      "loss": 5.1824,
      "mean_token_accuracy": 0.16846754848957063,
      "num_tokens": 23155966.0,
      "step": 10570
    },
    {
      "entropy": 5.50768814086914,
      "epoch": 1.0495236204843192,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004898487319928322,
      "loss": 5.1788,
      "mean_token_accuracy": 0.16355437636375428,
      "num_tokens": 23165896.0,
      "step": 10575
    },
    {
      "entropy": 5.493472671508789,
      "epoch": 1.0500198491464867,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004898382068324392,
      "loss": 5.1764,
      "mean_token_accuracy": 0.16505965292453767,
      "num_tokens": 23177821.0,
      "step": 10580
    },
    {
      "entropy": 5.52282304763794,
      "epoch": 1.0505160778086542,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004898276763445215,
      "loss": 5.2375,
      "mean_token_accuracy": 0.16412696093320847,
      "num_tokens": 23188916.0,
      "step": 10585
    },
    {
      "entropy": 5.607707548141479,
      "epoch": 1.0510123064708217,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004898171405293404,
      "loss": 5.2358,
      "mean_token_accuracy": 0.16866102665662766,
      "num_tokens": 23200490.0,
      "step": 10590
    },
    {
      "entropy": 5.484441757202148,
      "epoch": 1.0515085351329894,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004898065993871571,
      "loss": 5.1784,
      "mean_token_accuracy": 0.1647914916276932,
      "num_tokens": 23212991.0,
      "step": 10595
    },
    {
      "entropy": 5.551609706878662,
      "epoch": 1.0520047637951568,
      "grad_norm": 0.9375,
      "learning_rate": 0.000489796052918233,
      "loss": 5.2828,
      "mean_token_accuracy": 0.15930335968732834,
      "num_tokens": 23223361.0,
      "step": 10600
    },
    {
      "entropy": 5.553236389160157,
      "epoch": 1.0525009924573243,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004897855011228295,
      "loss": 5.1785,
      "mean_token_accuracy": 0.16796258538961412,
      "num_tokens": 23233765.0,
      "step": 10605
    },
    {
      "entropy": 5.492867755889892,
      "epoch": 1.0529972211194918,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004897749440012085,
      "loss": 5.1303,
      "mean_token_accuracy": 0.17219723612070084,
      "num_tokens": 23246562.0,
      "step": 10610
    },
    {
      "entropy": 5.469007396697998,
      "epoch": 1.0534934497816595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004897643815536316,
      "loss": 5.1474,
      "mean_token_accuracy": 0.1698348864912987,
      "num_tokens": 23256648.0,
      "step": 10615
    },
    {
      "entropy": 5.5521628856658936,
      "epoch": 1.053989678443827,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000489753813780361,
      "loss": 5.2127,
      "mean_token_accuracy": 0.16471905261278152,
      "num_tokens": 23267428.0,
      "step": 10620
    },
    {
      "entropy": 5.521107912063599,
      "epoch": 1.0544859071059944,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004897432406816585,
      "loss": 5.1463,
      "mean_token_accuracy": 0.16491916477680207,
      "num_tokens": 23278440.0,
      "step": 10625
    },
    {
      "entropy": 5.505084609985351,
      "epoch": 1.054982135768162,
      "grad_norm": 0.875,
      "learning_rate": 0.0004897326622577863,
      "loss": 5.1963,
      "mean_token_accuracy": 0.16055031418800353,
      "num_tokens": 23290571.0,
      "step": 10630
    },
    {
      "entropy": 5.5806097984313965,
      "epoch": 1.0554783644303294,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004897220785090067,
      "loss": 5.2633,
      "mean_token_accuracy": 0.16173502206802368,
      "num_tokens": 23301634.0,
      "step": 10635
    },
    {
      "entropy": 5.478116703033447,
      "epoch": 1.055974593092497,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004897114894355824,
      "loss": 5.1385,
      "mean_token_accuracy": 0.1717913791537285,
      "num_tokens": 23313125.0,
      "step": 10640
    },
    {
      "entropy": 5.479493474960327,
      "epoch": 1.0564708217546646,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004897008950377758,
      "loss": 5.1635,
      "mean_token_accuracy": 0.16100295931100844,
      "num_tokens": 23323610.0,
      "step": 10645
    },
    {
      "entropy": 5.451137685775757,
      "epoch": 1.056967050416832,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004896902953158497,
      "loss": 5.1394,
      "mean_token_accuracy": 0.1681753695011139,
      "num_tokens": 23333652.0,
      "step": 10650
    },
    {
      "entropy": 5.476093864440918,
      "epoch": 1.0574632790789995,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004896796902700667,
      "loss": 5.1827,
      "mean_token_accuracy": 0.16579088419675828,
      "num_tokens": 23345954.0,
      "step": 10655
    },
    {
      "entropy": 5.472770929336548,
      "epoch": 1.0579595077411672,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004896690799006901,
      "loss": 5.1541,
      "mean_token_accuracy": 0.1691794514656067,
      "num_tokens": 23357052.0,
      "step": 10660
    },
    {
      "entropy": 5.495822286605835,
      "epoch": 1.0584557364033347,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004896584642079828,
      "loss": 5.1812,
      "mean_token_accuracy": 0.17066397219896318,
      "num_tokens": 23368837.0,
      "step": 10665
    },
    {
      "entropy": 5.489329147338867,
      "epoch": 1.0589519650655022,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004896478431922083,
      "loss": 5.1548,
      "mean_token_accuracy": 0.16938245445489883,
      "num_tokens": 23378608.0,
      "step": 10670
    },
    {
      "entropy": 5.544188547134399,
      "epoch": 1.0594481937276696,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004896372168536297,
      "loss": 5.2215,
      "mean_token_accuracy": 0.16721830517053604,
      "num_tokens": 23390094.0,
      "step": 10675
    },
    {
      "entropy": 5.45720853805542,
      "epoch": 1.0599444223898373,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004896265851925107,
      "loss": 5.0931,
      "mean_token_accuracy": 0.1752452790737152,
      "num_tokens": 23399991.0,
      "step": 10680
    },
    {
      "entropy": 5.539272975921631,
      "epoch": 1.0604406510520048,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004896159482091148,
      "loss": 5.2897,
      "mean_token_accuracy": 0.15691623836755753,
      "num_tokens": 23410705.0,
      "step": 10685
    },
    {
      "entropy": 5.461822128295898,
      "epoch": 1.0609368797141723,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004896053059037058,
      "loss": 5.0726,
      "mean_token_accuracy": 0.18398592472076417,
      "num_tokens": 23421939.0,
      "step": 10690
    },
    {
      "entropy": 5.515606451034546,
      "epoch": 1.0614331083763398,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004895946582765477,
      "loss": 5.2103,
      "mean_token_accuracy": 0.16736398935317992,
      "num_tokens": 23433252.0,
      "step": 10695
    },
    {
      "entropy": 5.613071775436401,
      "epoch": 1.0619293370385074,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004895840053279045,
      "loss": 5.2772,
      "mean_token_accuracy": 0.15813700258731841,
      "num_tokens": 23445582.0,
      "step": 10700
    },
    {
      "entropy": 5.571057605743408,
      "epoch": 1.062425565700675,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004895733470580404,
      "loss": 5.2339,
      "mean_token_accuracy": 0.1600451797246933,
      "num_tokens": 23456360.0,
      "step": 10705
    },
    {
      "entropy": 5.524703931808472,
      "epoch": 1.0629217943628424,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004895626834672196,
      "loss": 5.1241,
      "mean_token_accuracy": 0.16391288936138154,
      "num_tokens": 23466576.0,
      "step": 10710
    },
    {
      "entropy": 5.421665954589844,
      "epoch": 1.0634180230250099,
      "grad_norm": 1.0,
      "learning_rate": 0.0004895520145557066,
      "loss": 5.0801,
      "mean_token_accuracy": 0.17319126725196837,
      "num_tokens": 23476200.0,
      "step": 10715
    },
    {
      "entropy": 5.533837461471558,
      "epoch": 1.0639142516871773,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004895413403237659,
      "loss": 5.2805,
      "mean_token_accuracy": 0.16643560528755189,
      "num_tokens": 23487776.0,
      "step": 10720
    },
    {
      "entropy": 5.519957256317139,
      "epoch": 1.064410480349345,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004895306607716624,
      "loss": 5.1327,
      "mean_token_accuracy": 0.1732277974486351,
      "num_tokens": 23497517.0,
      "step": 10725
    },
    {
      "entropy": 5.5163506984710695,
      "epoch": 1.0649067090115125,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004895199758996607,
      "loss": 5.1638,
      "mean_token_accuracy": 0.17083370983600615,
      "num_tokens": 23507436.0,
      "step": 10730
    },
    {
      "entropy": 5.479664707183838,
      "epoch": 1.06540293767368,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004895092857080258,
      "loss": 5.1094,
      "mean_token_accuracy": 0.1597520738840103,
      "num_tokens": 23517170.0,
      "step": 10735
    },
    {
      "entropy": 5.479441976547241,
      "epoch": 1.0658991663358475,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004894985901970229,
      "loss": 5.1174,
      "mean_token_accuracy": 0.173440021276474,
      "num_tokens": 23527329.0,
      "step": 10740
    },
    {
      "entropy": 5.561740398406982,
      "epoch": 1.0663953949980152,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004894878893669171,
      "loss": 5.2226,
      "mean_token_accuracy": 0.16073554754257202,
      "num_tokens": 23538341.0,
      "step": 10745
    },
    {
      "entropy": 5.505781459808349,
      "epoch": 1.0668916236601826,
      "grad_norm": 0.875,
      "learning_rate": 0.0004894771832179739,
      "loss": 5.1448,
      "mean_token_accuracy": 0.16851148307323455,
      "num_tokens": 23549522.0,
      "step": 10750
    },
    {
      "entropy": 5.569004535675049,
      "epoch": 1.0673878523223501,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004894664717504587,
      "loss": 5.186,
      "mean_token_accuracy": 0.1654226675629616,
      "num_tokens": 23559274.0,
      "step": 10755
    },
    {
      "entropy": 5.525196743011475,
      "epoch": 1.0678840809845176,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004894557549646371,
      "loss": 5.1539,
      "mean_token_accuracy": 0.17346994131803511,
      "num_tokens": 23570852.0,
      "step": 10760
    },
    {
      "entropy": 5.4818190574646,
      "epoch": 1.0683803096466853,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004894450328607749,
      "loss": 5.1488,
      "mean_token_accuracy": 0.16973407864570617,
      "num_tokens": 23582918.0,
      "step": 10765
    },
    {
      "entropy": 5.473199987411499,
      "epoch": 1.0688765383088528,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004894343054391379,
      "loss": 5.1838,
      "mean_token_accuracy": 0.1745926633477211,
      "num_tokens": 23594177.0,
      "step": 10770
    },
    {
      "entropy": 5.504260063171387,
      "epoch": 1.0693727669710202,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004894235726999924,
      "loss": 5.1418,
      "mean_token_accuracy": 0.17183480858802797,
      "num_tokens": 23605901.0,
      "step": 10775
    },
    {
      "entropy": 5.4763895034790036,
      "epoch": 1.0698689956331877,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004894128346436042,
      "loss": 5.1963,
      "mean_token_accuracy": 0.1675628513097763,
      "num_tokens": 23617354.0,
      "step": 10780
    },
    {
      "entropy": 5.522255086898804,
      "epoch": 1.0703652242953554,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004894020912702397,
      "loss": 5.221,
      "mean_token_accuracy": 0.1643326163291931,
      "num_tokens": 23627603.0,
      "step": 10785
    },
    {
      "entropy": 5.440590858459473,
      "epoch": 1.0708614529575229,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004893913425801653,
      "loss": 5.0436,
      "mean_token_accuracy": 0.17594035416841508,
      "num_tokens": 23637280.0,
      "step": 10790
    },
    {
      "entropy": 5.517892932891845,
      "epoch": 1.0713576816196904,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004893805885736477,
      "loss": 5.2475,
      "mean_token_accuracy": 0.16413678079843522,
      "num_tokens": 23649326.0,
      "step": 10795
    },
    {
      "entropy": 5.568311405181885,
      "epoch": 1.0718539102818578,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004893698292509534,
      "loss": 5.2488,
      "mean_token_accuracy": 0.1645419031381607,
      "num_tokens": 23659561.0,
      "step": 10800
    },
    {
      "entropy": 5.518482828140259,
      "epoch": 1.0723501389440253,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004893590646123492,
      "loss": 5.1611,
      "mean_token_accuracy": 0.1690828889608383,
      "num_tokens": 23673206.0,
      "step": 10805
    },
    {
      "entropy": 5.491669321060181,
      "epoch": 1.072846367606193,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004893482946581022,
      "loss": 5.1949,
      "mean_token_accuracy": 0.16448067724704743,
      "num_tokens": 23683920.0,
      "step": 10810
    },
    {
      "entropy": 5.461504507064819,
      "epoch": 1.0733425962683605,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004893375193884793,
      "loss": 5.1147,
      "mean_token_accuracy": 0.17293360084295273,
      "num_tokens": 23693990.0,
      "step": 10815
    },
    {
      "entropy": 5.5457727909088135,
      "epoch": 1.073838824930528,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004893267388037478,
      "loss": 5.2244,
      "mean_token_accuracy": 0.16552564948797227,
      "num_tokens": 23704870.0,
      "step": 10820
    },
    {
      "entropy": 5.45700101852417,
      "epoch": 1.0743350535926954,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000489315952904175,
      "loss": 5.2099,
      "mean_token_accuracy": 0.16702677607536315,
      "num_tokens": 23715159.0,
      "step": 10825
    },
    {
      "entropy": 5.489004421234131,
      "epoch": 1.0748312822548631,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004893051616900283,
      "loss": 5.138,
      "mean_token_accuracy": 0.16661283373832703,
      "num_tokens": 23725252.0,
      "step": 10830
    },
    {
      "entropy": 5.484555721282959,
      "epoch": 1.0753275109170306,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004892943651615756,
      "loss": 5.1542,
      "mean_token_accuracy": 0.17478515952825546,
      "num_tokens": 23736472.0,
      "step": 10835
    },
    {
      "entropy": 5.500439023971557,
      "epoch": 1.075823739579198,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004892835633190843,
      "loss": 5.0729,
      "mean_token_accuracy": 0.16555219292640685,
      "num_tokens": 23747565.0,
      "step": 10840
    },
    {
      "entropy": 5.488166046142578,
      "epoch": 1.0763199682413656,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004892727561628223,
      "loss": 5.1693,
      "mean_token_accuracy": 0.16721499860286712,
      "num_tokens": 23758201.0,
      "step": 10845
    },
    {
      "entropy": 5.543577861785889,
      "epoch": 1.0768161969035333,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004892619436930577,
      "loss": 5.2183,
      "mean_token_accuracy": 0.1680361270904541,
      "num_tokens": 23769087.0,
      "step": 10850
    },
    {
      "entropy": 5.541718244552612,
      "epoch": 1.0773124255657007,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004892511259100585,
      "loss": 5.2133,
      "mean_token_accuracy": 0.16787203550338745,
      "num_tokens": 23780598.0,
      "step": 10855
    },
    {
      "entropy": 5.514758491516114,
      "epoch": 1.0778086542278682,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004892403028140932,
      "loss": 5.1403,
      "mean_token_accuracy": 0.16158319115638733,
      "num_tokens": 23791449.0,
      "step": 10860
    },
    {
      "entropy": 5.4929993629455565,
      "epoch": 1.0783048828900357,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048922947440543,
      "loss": 5.1332,
      "mean_token_accuracy": 0.17096055895090104,
      "num_tokens": 23803071.0,
      "step": 10865
    },
    {
      "entropy": 5.533701801300049,
      "epoch": 1.0788011115522032,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004892186406843374,
      "loss": 5.2057,
      "mean_token_accuracy": 0.16061035692691802,
      "num_tokens": 23813372.0,
      "step": 10870
    },
    {
      "entropy": 5.491838121414185,
      "epoch": 1.0792973402143708,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004892078016510841,
      "loss": 5.1706,
      "mean_token_accuracy": 0.16986196339130402,
      "num_tokens": 23824979.0,
      "step": 10875
    },
    {
      "entropy": 5.502412128448486,
      "epoch": 1.0797935688765383,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000489196957305939,
      "loss": 5.1622,
      "mean_token_accuracy": 0.1653856858611107,
      "num_tokens": 23836335.0,
      "step": 10880
    },
    {
      "entropy": 5.4904509544372555,
      "epoch": 1.0802897975387058,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004891861076491709,
      "loss": 5.1567,
      "mean_token_accuracy": 0.17029475420713425,
      "num_tokens": 23848147.0,
      "step": 10885
    },
    {
      "entropy": 5.451874876022339,
      "epoch": 1.0807860262008733,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004891752526810488,
      "loss": 5.1069,
      "mean_token_accuracy": 0.17364071756601335,
      "num_tokens": 23859502.0,
      "step": 10890
    },
    {
      "entropy": 5.522591018676758,
      "epoch": 1.081282254863041,
      "grad_norm": 0.84375,
      "learning_rate": 0.000489164392401842,
      "loss": 5.2518,
      "mean_token_accuracy": 0.16657178103923798,
      "num_tokens": 23870433.0,
      "step": 10895
    },
    {
      "entropy": 5.476272773742676,
      "epoch": 1.0817784835252084,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004891535268118196,
      "loss": 5.0635,
      "mean_token_accuracy": 0.1754612922668457,
      "num_tokens": 23879581.0,
      "step": 10900
    },
    {
      "entropy": 5.424292087554932,
      "epoch": 1.082274712187376,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004891426559112514,
      "loss": 5.1412,
      "mean_token_accuracy": 0.17278343439102173,
      "num_tokens": 23890889.0,
      "step": 10905
    },
    {
      "entropy": 5.4880101680755615,
      "epoch": 1.0827709408495434,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004891317797004066,
      "loss": 5.1348,
      "mean_token_accuracy": 0.17125614285469054,
      "num_tokens": 23901577.0,
      "step": 10910
    },
    {
      "entropy": 5.470410585403442,
      "epoch": 1.083267169511711,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004891208981795552,
      "loss": 5.2133,
      "mean_token_accuracy": 0.16936561316251755,
      "num_tokens": 23912698.0,
      "step": 10915
    },
    {
      "entropy": 5.417942237854004,
      "epoch": 1.0837633981738786,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004891100113489669,
      "loss": 5.0855,
      "mean_token_accuracy": 0.1729748159646988,
      "num_tokens": 23923056.0,
      "step": 10920
    },
    {
      "entropy": 5.475066804885865,
      "epoch": 1.084259626836046,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004890991192089116,
      "loss": 5.2276,
      "mean_token_accuracy": 0.16701621562242508,
      "num_tokens": 23933998.0,
      "step": 10925
    },
    {
      "entropy": 5.541318655014038,
      "epoch": 1.0847558554982135,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004890882217596595,
      "loss": 5.1268,
      "mean_token_accuracy": 0.1713723510503769,
      "num_tokens": 23943980.0,
      "step": 10930
    },
    {
      "entropy": 5.499484539031982,
      "epoch": 1.085252084160381,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004890773190014809,
      "loss": 5.193,
      "mean_token_accuracy": 0.16638527512550355,
      "num_tokens": 23955646.0,
      "step": 10935
    },
    {
      "entropy": 5.456666898727417,
      "epoch": 1.0857483128225487,
      "grad_norm": 0.9375,
      "learning_rate": 0.000489066410934646,
      "loss": 5.1657,
      "mean_token_accuracy": 0.1730557292699814,
      "num_tokens": 23966381.0,
      "step": 10940
    },
    {
      "entropy": 5.5801537990570065,
      "epoch": 1.0862445414847162,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004890554975594255,
      "loss": 5.2391,
      "mean_token_accuracy": 0.15644838511943818,
      "num_tokens": 23978669.0,
      "step": 10945
    },
    {
      "entropy": 5.626119947433471,
      "epoch": 1.0867407701468836,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004890445788760898,
      "loss": 5.2926,
      "mean_token_accuracy": 0.16053656488656998,
      "num_tokens": 23989467.0,
      "step": 10950
    },
    {
      "entropy": 5.492971706390381,
      "epoch": 1.0872369988090511,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004890336548849098,
      "loss": 5.1128,
      "mean_token_accuracy": 0.17263050973415375,
      "num_tokens": 24000797.0,
      "step": 10955
    },
    {
      "entropy": 5.582439327239991,
      "epoch": 1.0877332274712188,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004890227255861564,
      "loss": 5.2621,
      "mean_token_accuracy": 0.1681663379073143,
      "num_tokens": 24013182.0,
      "step": 10960
    },
    {
      "entropy": 5.5321564197540285,
      "epoch": 1.0882294561333863,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004890117909801005,
      "loss": 5.1396,
      "mean_token_accuracy": 0.17512231767177583,
      "num_tokens": 24024005.0,
      "step": 10965
    },
    {
      "entropy": 5.502088308334351,
      "epoch": 1.0887256847955538,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004890008510670134,
      "loss": 5.1377,
      "mean_token_accuracy": 0.1663498252630234,
      "num_tokens": 24034524.0,
      "step": 10970
    },
    {
      "entropy": 5.533322477340699,
      "epoch": 1.0892219134577212,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004889899058471663,
      "loss": 5.2013,
      "mean_token_accuracy": 0.16822793036699296,
      "num_tokens": 24045944.0,
      "step": 10975
    },
    {
      "entropy": 5.575636863708496,
      "epoch": 1.089718142119889,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004889789553208306,
      "loss": 5.268,
      "mean_token_accuracy": 0.1660327509045601,
      "num_tokens": 24058612.0,
      "step": 10980
    },
    {
      "entropy": 5.520413494110107,
      "epoch": 1.0902143707820564,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000488967999488278,
      "loss": 5.1379,
      "mean_token_accuracy": 0.17078853994607926,
      "num_tokens": 24068912.0,
      "step": 10985
    },
    {
      "entropy": 5.419206237792968,
      "epoch": 1.0907105994442239,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048895703834978,
      "loss": 5.1451,
      "mean_token_accuracy": 0.1665818512439728,
      "num_tokens": 24079879.0,
      "step": 10990
    },
    {
      "entropy": 5.503745603561401,
      "epoch": 1.0912068281063914,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004889460719056085,
      "loss": 5.1674,
      "mean_token_accuracy": 0.16840779334306716,
      "num_tokens": 24091268.0,
      "step": 10995
    },
    {
      "entropy": 5.518901920318603,
      "epoch": 1.091703056768559,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004889351001560354,
      "loss": 5.1914,
      "mean_token_accuracy": 0.16596200168132783,
      "num_tokens": 24101428.0,
      "step": 11000
    },
    {
      "entropy": 5.464192247390747,
      "epoch": 1.0921992854307265,
      "grad_norm": 0.875,
      "learning_rate": 0.0004889241231013329,
      "loss": 5.1585,
      "mean_token_accuracy": 0.1704880937933922,
      "num_tokens": 24112920.0,
      "step": 11005
    },
    {
      "entropy": 5.458786821365356,
      "epoch": 1.092695514092894,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000488913140741773,
      "loss": 5.0893,
      "mean_token_accuracy": 0.16703478693962098,
      "num_tokens": 24125731.0,
      "step": 11010
    },
    {
      "entropy": 5.479241228103637,
      "epoch": 1.0931917427550615,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004889021530776282,
      "loss": 5.1051,
      "mean_token_accuracy": 0.16768449544906616,
      "num_tokens": 24136606.0,
      "step": 11015
    },
    {
      "entropy": 5.550204944610596,
      "epoch": 1.093687971417229,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004888911601091709,
      "loss": 5.125,
      "mean_token_accuracy": 0.1734997034072876,
      "num_tokens": 24146783.0,
      "step": 11020
    },
    {
      "entropy": 5.499466609954834,
      "epoch": 1.0941842000793967,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004888801618366737,
      "loss": 5.226,
      "mean_token_accuracy": 0.1696036234498024,
      "num_tokens": 24157412.0,
      "step": 11025
    },
    {
      "entropy": 5.516845607757569,
      "epoch": 1.0946804287415641,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004888691582604095,
      "loss": 5.209,
      "mean_token_accuracy": 0.16557887345552444,
      "num_tokens": 24168077.0,
      "step": 11030
    },
    {
      "entropy": 5.553576278686523,
      "epoch": 1.0951766574037316,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000488858149380651,
      "loss": 5.2585,
      "mean_token_accuracy": 0.16936648190021514,
      "num_tokens": 24178913.0,
      "step": 11035
    },
    {
      "entropy": 5.5928089141845705,
      "epoch": 1.095672886065899,
      "grad_norm": 0.96875,
      "learning_rate": 0.000488847135197671,
      "loss": 5.3005,
      "mean_token_accuracy": 0.158344267308712,
      "num_tokens": 24190356.0,
      "step": 11040
    },
    {
      "entropy": 5.57515230178833,
      "epoch": 1.0961691147280668,
      "grad_norm": 1.046875,
      "learning_rate": 0.000488836115711743,
      "loss": 5.2473,
      "mean_token_accuracy": 0.161076557636261,
      "num_tokens": 24199772.0,
      "step": 11045
    },
    {
      "entropy": 5.540526580810547,
      "epoch": 1.0966653433902342,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004888250909231401,
      "loss": 5.1557,
      "mean_token_accuracy": 0.16764594614505768,
      "num_tokens": 24211807.0,
      "step": 11050
    },
    {
      "entropy": 5.476969623565674,
      "epoch": 1.0971615720524017,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004888140608321357,
      "loss": 5.1485,
      "mean_token_accuracy": 0.17423913180828093,
      "num_tokens": 24223156.0,
      "step": 11055
    },
    {
      "entropy": 5.567568063735962,
      "epoch": 1.0976578007145692,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004888030254390033,
      "loss": 5.2115,
      "mean_token_accuracy": 0.16370397061109543,
      "num_tokens": 24234352.0,
      "step": 11060
    },
    {
      "entropy": 5.516723585128784,
      "epoch": 1.098154029376737,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004887919847440166,
      "loss": 5.1298,
      "mean_token_accuracy": 0.1740872636437416,
      "num_tokens": 24244137.0,
      "step": 11065
    },
    {
      "entropy": 5.463329601287842,
      "epoch": 1.0986502580389044,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004887809387474494,
      "loss": 5.1369,
      "mean_token_accuracy": 0.16834257394075394,
      "num_tokens": 24253250.0,
      "step": 11070
    },
    {
      "entropy": 5.537700366973877,
      "epoch": 1.0991464867010718,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004887698874495755,
      "loss": 5.252,
      "mean_token_accuracy": 0.16806362569332123,
      "num_tokens": 24264551.0,
      "step": 11075
    },
    {
      "entropy": 5.508148956298828,
      "epoch": 1.0996427153632393,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000488758830850669,
      "loss": 5.0704,
      "mean_token_accuracy": 0.17682563811540603,
      "num_tokens": 24275522.0,
      "step": 11080
    },
    {
      "entropy": 5.475867557525635,
      "epoch": 1.100138944025407,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004887477689510042,
      "loss": 5.2243,
      "mean_token_accuracy": 0.16861492395401,
      "num_tokens": 24286333.0,
      "step": 11085
    },
    {
      "entropy": 5.475257730484008,
      "epoch": 1.1006351726875745,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004887367017508553,
      "loss": 5.2385,
      "mean_token_accuracy": 0.16946501284837723,
      "num_tokens": 24297627.0,
      "step": 11090
    },
    {
      "entropy": 5.495471620559693,
      "epoch": 1.101131401349742,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004887256292504967,
      "loss": 5.1766,
      "mean_token_accuracy": 0.16979029923677444,
      "num_tokens": 24308246.0,
      "step": 11095
    },
    {
      "entropy": 5.541315174102783,
      "epoch": 1.1016276300119094,
      "grad_norm": 1.078125,
      "learning_rate": 0.000488714551450203,
      "loss": 5.2241,
      "mean_token_accuracy": 0.162785604596138,
      "num_tokens": 24317718.0,
      "step": 11100
    },
    {
      "entropy": 5.501359224319458,
      "epoch": 1.102123858674077,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000488703468350249,
      "loss": 5.1548,
      "mean_token_accuracy": 0.16744553297758102,
      "num_tokens": 24329242.0,
      "step": 11105
    },
    {
      "entropy": 5.616970062255859,
      "epoch": 1.1026200873362446,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004886923799509095,
      "loss": 5.3133,
      "mean_token_accuracy": 0.16358482390642165,
      "num_tokens": 24340146.0,
      "step": 11110
    },
    {
      "entropy": 5.564360427856445,
      "epoch": 1.103116315998412,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004886812862524595,
      "loss": 5.2128,
      "mean_token_accuracy": 0.16251914203166962,
      "num_tokens": 24349370.0,
      "step": 11115
    },
    {
      "entropy": 5.573534393310547,
      "epoch": 1.1036125446605796,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004886701872551739,
      "loss": 5.1599,
      "mean_token_accuracy": 0.1653120204806328,
      "num_tokens": 24359770.0,
      "step": 11120
    },
    {
      "entropy": 5.5279176235198975,
      "epoch": 1.104108773322747,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004886590829593281,
      "loss": 5.2082,
      "mean_token_accuracy": 0.1692720517516136,
      "num_tokens": 24370974.0,
      "step": 11125
    },
    {
      "entropy": 5.529518032073975,
      "epoch": 1.1046050019849147,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004886479733651974,
      "loss": 5.205,
      "mean_token_accuracy": 0.167129485309124,
      "num_tokens": 24381771.0,
      "step": 11130
    },
    {
      "entropy": 5.481357479095459,
      "epoch": 1.1051012306470822,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004886368584730575,
      "loss": 5.172,
      "mean_token_accuracy": 0.16281089633703233,
      "num_tokens": 24393067.0,
      "step": 11135
    },
    {
      "entropy": 5.505965518951416,
      "epoch": 1.1055974593092497,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004886257382831837,
      "loss": 5.2394,
      "mean_token_accuracy": 0.1625291511416435,
      "num_tokens": 24402866.0,
      "step": 11140
    },
    {
      "entropy": 5.577435398101807,
      "epoch": 1.1060936879714172,
      "grad_norm": 0.953125,
      "learning_rate": 0.000488614612795852,
      "loss": 5.1683,
      "mean_token_accuracy": 0.169407357275486,
      "num_tokens": 24412956.0,
      "step": 11145
    },
    {
      "entropy": 5.573137950897217,
      "epoch": 1.1065899166335849,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004886034820113382,
      "loss": 5.2377,
      "mean_token_accuracy": 0.16365133225917816,
      "num_tokens": 24424650.0,
      "step": 11150
    },
    {
      "entropy": 5.555062007904053,
      "epoch": 1.1070861452957523,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004885923459299182,
      "loss": 5.2526,
      "mean_token_accuracy": 0.16688985675573348,
      "num_tokens": 24436541.0,
      "step": 11155
    },
    {
      "entropy": 5.571127605438233,
      "epoch": 1.1075823739579198,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004885812045518685,
      "loss": 5.2387,
      "mean_token_accuracy": 0.16248765736818313,
      "num_tokens": 24447588.0,
      "step": 11160
    },
    {
      "entropy": 5.570699787139892,
      "epoch": 1.1080786026200873,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000488570057877465,
      "loss": 5.1487,
      "mean_token_accuracy": 0.1717093899846077,
      "num_tokens": 24458854.0,
      "step": 11165
    },
    {
      "entropy": 5.5340282917022705,
      "epoch": 1.108574831282255,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004885589059069842,
      "loss": 5.2161,
      "mean_token_accuracy": 0.16735346764326095,
      "num_tokens": 24469427.0,
      "step": 11170
    },
    {
      "entropy": 5.455790281295776,
      "epoch": 1.1090710599444225,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004885477486407028,
      "loss": 5.2115,
      "mean_token_accuracy": 0.16059117913246154,
      "num_tokens": 24479352.0,
      "step": 11175
    },
    {
      "entropy": 5.601591444015503,
      "epoch": 1.10956728860659,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004885365860788974,
      "loss": 5.2,
      "mean_token_accuracy": 0.16488547176122664,
      "num_tokens": 24489580.0,
      "step": 11180
    },
    {
      "entropy": 5.519624948501587,
      "epoch": 1.1100635172687574,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004885254182218448,
      "loss": 5.1626,
      "mean_token_accuracy": 0.1684727668762207,
      "num_tokens": 24501141.0,
      "step": 11185
    },
    {
      "entropy": 5.479819393157959,
      "epoch": 1.1105597459309249,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004885142450698219,
      "loss": 5.1642,
      "mean_token_accuracy": 0.16967890560626983,
      "num_tokens": 24512079.0,
      "step": 11190
    },
    {
      "entropy": 5.552234888076782,
      "epoch": 1.1110559745930926,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004885030666231057,
      "loss": 5.2164,
      "mean_token_accuracy": 0.1667739197611809,
      "num_tokens": 24521839.0,
      "step": 11195
    },
    {
      "entropy": 5.551833391189575,
      "epoch": 1.11155220325526,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004884918828819738,
      "loss": 5.259,
      "mean_token_accuracy": 0.16346664428710939,
      "num_tokens": 24532704.0,
      "step": 11200
    },
    {
      "entropy": 5.524718379974365,
      "epoch": 1.1120484319174275,
      "grad_norm": 0.90625,
      "learning_rate": 0.000488480693846703,
      "loss": 5.1619,
      "mean_token_accuracy": 0.1717764914035797,
      "num_tokens": 24543790.0,
      "step": 11205
    },
    {
      "entropy": 5.499316644668579,
      "epoch": 1.112544660579595,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004884694995175711,
      "loss": 5.2149,
      "mean_token_accuracy": 0.16575611680746077,
      "num_tokens": 24554765.0,
      "step": 11210
    },
    {
      "entropy": 5.64555983543396,
      "epoch": 1.1130408892417627,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004884582998948555,
      "loss": 5.2889,
      "mean_token_accuracy": 0.15843822062015533,
      "num_tokens": 24565888.0,
      "step": 11215
    },
    {
      "entropy": 5.487444019317627,
      "epoch": 1.1135371179039302,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004884470949788342,
      "loss": 5.1487,
      "mean_token_accuracy": 0.1711251765489578,
      "num_tokens": 24575482.0,
      "step": 11220
    },
    {
      "entropy": 5.562432670593262,
      "epoch": 1.1140333465660976,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004884358847697847,
      "loss": 5.2724,
      "mean_token_accuracy": 0.15471834540367127,
      "num_tokens": 24588249.0,
      "step": 11225
    },
    {
      "entropy": 5.59770097732544,
      "epoch": 1.1145295752282651,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004884246692679852,
      "loss": 5.2105,
      "mean_token_accuracy": 0.1702192947268486,
      "num_tokens": 24599546.0,
      "step": 11230
    },
    {
      "entropy": 5.517202663421631,
      "epoch": 1.1150258038904326,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004884134484737139,
      "loss": 5.1551,
      "mean_token_accuracy": 0.16802441328763962,
      "num_tokens": 24611777.0,
      "step": 11235
    },
    {
      "entropy": 5.498545980453491,
      "epoch": 1.1155220325526003,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004884022223872489,
      "loss": 5.1362,
      "mean_token_accuracy": 0.16747527122497557,
      "num_tokens": 24623675.0,
      "step": 11240
    },
    {
      "entropy": 5.52301025390625,
      "epoch": 1.1160182612147678,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004883909910088686,
      "loss": 5.1652,
      "mean_token_accuracy": 0.16839252710342406,
      "num_tokens": 24633929.0,
      "step": 11245
    },
    {
      "entropy": 5.485877895355225,
      "epoch": 1.1165144898769352,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004883797543388515,
      "loss": 5.127,
      "mean_token_accuracy": 0.1698381707072258,
      "num_tokens": 24642800.0,
      "step": 11250
    },
    {
      "entropy": 5.456842374801636,
      "epoch": 1.1170107185391027,
      "grad_norm": 1.0,
      "learning_rate": 0.0004883685123774764,
      "loss": 5.1215,
      "mean_token_accuracy": 0.17325924783945085,
      "num_tokens": 24653780.0,
      "step": 11255
    },
    {
      "entropy": 5.472145652770996,
      "epoch": 1.1175069472012704,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004883572651250218,
      "loss": 5.1606,
      "mean_token_accuracy": 0.16740617752075196,
      "num_tokens": 24663912.0,
      "step": 11260
    },
    {
      "entropy": 5.536586141586303,
      "epoch": 1.118003175863438,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004883460125817669,
      "loss": 5.2486,
      "mean_token_accuracy": 0.1640252113342285,
      "num_tokens": 24675222.0,
      "step": 11265
    },
    {
      "entropy": 5.563428974151611,
      "epoch": 1.1184994045256054,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004883347547479905,
      "loss": 5.253,
      "mean_token_accuracy": 0.16681550741195678,
      "num_tokens": 24686697.0,
      "step": 11270
    },
    {
      "entropy": 5.571768045425415,
      "epoch": 1.1189956331877728,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000488323491623972,
      "loss": 5.1623,
      "mean_token_accuracy": 0.17624045610427858,
      "num_tokens": 24697067.0,
      "step": 11275
    },
    {
      "entropy": 5.548267364501953,
      "epoch": 1.1194918618499405,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004883122232099905,
      "loss": 5.1754,
      "mean_token_accuracy": 0.16944562941789626,
      "num_tokens": 24707989.0,
      "step": 11280
    },
    {
      "entropy": 5.431782197952271,
      "epoch": 1.119988090512108,
      "grad_norm": 0.875,
      "learning_rate": 0.0004883009495063256,
      "loss": 5.1139,
      "mean_token_accuracy": 0.17899062782526015,
      "num_tokens": 24718826.0,
      "step": 11285
    },
    {
      "entropy": 5.541382598876953,
      "epoch": 1.1204843191742755,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004882896705132566,
      "loss": 5.2302,
      "mean_token_accuracy": 0.1619640827178955,
      "num_tokens": 24729913.0,
      "step": 11290
    },
    {
      "entropy": 5.454885005950928,
      "epoch": 1.120980547836443,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048827838623106355,
      "loss": 5.0974,
      "mean_token_accuracy": 0.1714236170053482,
      "num_tokens": 24741621.0,
      "step": 11295
    },
    {
      "entropy": 5.5092963695526125,
      "epoch": 1.1214767764986107,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000488267096660026,
      "loss": 5.147,
      "mean_token_accuracy": 0.16761314421892165,
      "num_tokens": 24753427.0,
      "step": 11300
    },
    {
      "entropy": 5.549616384506225,
      "epoch": 1.1219730051607781,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048825580180042406,
      "loss": 5.1816,
      "mean_token_accuracy": 0.17004798501729965,
      "num_tokens": 24763580.0,
      "step": 11305
    },
    {
      "entropy": 5.481188058853149,
      "epoch": 1.1224692338229456,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004882445016525377,
      "loss": 5.1681,
      "mean_token_accuracy": 0.16722124814987183,
      "num_tokens": 24773955.0,
      "step": 11310
    },
    {
      "entropy": 5.511306381225586,
      "epoch": 1.122965462485113,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048823319621664723,
      "loss": 5.0865,
      "mean_token_accuracy": 0.1783554509282112,
      "num_tokens": 24784039.0,
      "step": 11315
    },
    {
      "entropy": 5.583564329147339,
      "epoch": 1.1234616911472806,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048822188549303296,
      "loss": 5.2022,
      "mean_token_accuracy": 0.16035133898258208,
      "num_tokens": 24794630.0,
      "step": 11320
    },
    {
      "entropy": 5.557933139801025,
      "epoch": 1.1239579198094483,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004882105694819754,
      "loss": 5.2514,
      "mean_token_accuracy": 0.1608917698264122,
      "num_tokens": 24805625.0,
      "step": 11325
    },
    {
      "entropy": 5.555304145812988,
      "epoch": 1.1244541484716157,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004881992481837551,
      "loss": 5.1837,
      "mean_token_accuracy": 0.175361330807209,
      "num_tokens": 24817356.0,
      "step": 11330
    },
    {
      "entropy": 5.503820419311523,
      "epoch": 1.1249503771337832,
      "grad_norm": 0.875,
      "learning_rate": 0.0004881879215986529,
      "loss": 5.1497,
      "mean_token_accuracy": 0.16104721277952194,
      "num_tokens": 24828349.0,
      "step": 11335
    },
    {
      "entropy": 5.518275547027588,
      "epoch": 1.1254466057959507,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048817658972694967,
      "loss": 5.1913,
      "mean_token_accuracy": 0.1651931494474411,
      "num_tokens": 24840138.0,
      "step": 11340
    },
    {
      "entropy": 5.532198524475097,
      "epoch": 1.1259428344581184,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004881652525689263,
      "loss": 5.1866,
      "mean_token_accuracy": 0.16392576992511748,
      "num_tokens": 24851570.0,
      "step": 11345
    },
    {
      "entropy": 5.406888103485107,
      "epoch": 1.1264390631202859,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000488153910124864,
      "loss": 5.0813,
      "mean_token_accuracy": 0.17003836184740068,
      "num_tokens": 24861396.0,
      "step": 11350
    },
    {
      "entropy": 5.527017307281494,
      "epoch": 1.1269352917824533,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000488142562395044,
      "loss": 5.2067,
      "mean_token_accuracy": 0.16171378791332244,
      "num_tokens": 24873519.0,
      "step": 11355
    },
    {
      "entropy": 5.5367402076721195,
      "epoch": 1.1274315204446208,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00048813120937974774,
      "loss": 5.2196,
      "mean_token_accuracy": 0.16119163632392883,
      "num_tokens": 24886856.0,
      "step": 11360
    },
    {
      "entropy": 5.489326763153076,
      "epoch": 1.1279277491067885,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048811985107925684,
      "loss": 5.1744,
      "mean_token_accuracy": 0.16307065784931182,
      "num_tokens": 24897292.0,
      "step": 11365
    },
    {
      "entropy": 5.536104440689087,
      "epoch": 1.128423977768956,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048810848749385286,
      "loss": 5.1807,
      "mean_token_accuracy": 0.16587122082710265,
      "num_tokens": 24907757.0,
      "step": 11370
    },
    {
      "entropy": 5.453400373458862,
      "epoch": 1.1289202064311235,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048809711862381753,
      "loss": 5.1591,
      "mean_token_accuracy": 0.1707037642598152,
      "num_tokens": 24918071.0,
      "step": 11375
    },
    {
      "entropy": 5.48599762916565,
      "epoch": 1.129416435093291,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048808574446943274,
      "loss": 5.1332,
      "mean_token_accuracy": 0.17266378104686736,
      "num_tokens": 24927967.0,
      "step": 11380
    },
    {
      "entropy": 5.5505107879638675,
      "epoch": 1.1299126637554586,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004880743650309808,
      "loss": 5.207,
      "mean_token_accuracy": 0.17195654958486556,
      "num_tokens": 24939170.0,
      "step": 11385
    },
    {
      "entropy": 5.512000846862793,
      "epoch": 1.130408892417626,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00048806298030874373,
      "loss": 5.1922,
      "mean_token_accuracy": 0.1708056300878525,
      "num_tokens": 24950937.0,
      "step": 11390
    },
    {
      "entropy": 5.500658226013184,
      "epoch": 1.1309051210797936,
      "grad_norm": 0.81640625,
      "learning_rate": 0.00048805159030300383,
      "loss": 5.1834,
      "mean_token_accuracy": 0.16870294362306595,
      "num_tokens": 24962813.0,
      "step": 11395
    },
    {
      "entropy": 5.497263860702515,
      "epoch": 1.131401349741961,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048804019501404357,
      "loss": 5.0778,
      "mean_token_accuracy": 0.17271201312541962,
      "num_tokens": 24974192.0,
      "step": 11400
    },
    {
      "entropy": 5.461179494857788,
      "epoch": 1.1318975784041285,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048802879444214553,
      "loss": 5.0843,
      "mean_token_accuracy": 0.17591992020606995,
      "num_tokens": 24985353.0,
      "step": 11405
    },
    {
      "entropy": 5.498987674713135,
      "epoch": 1.1323938070662962,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00048801738858759243,
      "loss": 5.1904,
      "mean_token_accuracy": 0.17054355442523955,
      "num_tokens": 24997259.0,
      "step": 11410
    },
    {
      "entropy": 5.507398080825806,
      "epoch": 1.1328900357284637,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004880059774506671,
      "loss": 5.1777,
      "mean_token_accuracy": 0.16675892919301988,
      "num_tokens": 25008222.0,
      "step": 11415
    },
    {
      "entropy": 5.512610197067261,
      "epoch": 1.1333862643906312,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004879945610316525,
      "loss": 5.2189,
      "mean_token_accuracy": 0.16371117830276488,
      "num_tokens": 25020267.0,
      "step": 11420
    },
    {
      "entropy": 5.516874551773071,
      "epoch": 1.1338824930527986,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048798313933083175,
      "loss": 5.1886,
      "mean_token_accuracy": 0.1679120123386383,
      "num_tokens": 25031212.0,
      "step": 11425
    },
    {
      "entropy": 5.514549446105957,
      "epoch": 1.1343787217149663,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004879717123484882,
      "loss": 5.1548,
      "mean_token_accuracy": 0.16509698778390886,
      "num_tokens": 25042204.0,
      "step": 11430
    },
    {
      "entropy": 5.542244005203247,
      "epoch": 1.1348749503771338,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000487960280084905,
      "loss": 5.1812,
      "mean_token_accuracy": 0.1701316550374031,
      "num_tokens": 25052010.0,
      "step": 11435
    },
    {
      "entropy": 5.551565837860108,
      "epoch": 1.1353711790393013,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004879488425403657,
      "loss": 5.2296,
      "mean_token_accuracy": 0.161693225055933,
      "num_tokens": 25062209.0,
      "step": 11440
    },
    {
      "entropy": 5.504443120956421,
      "epoch": 1.1358674077014688,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048793739971515407,
      "loss": 5.1864,
      "mean_token_accuracy": 0.16541998088359833,
      "num_tokens": 25072758.0,
      "step": 11445
    },
    {
      "entropy": 5.549315118789673,
      "epoch": 1.1363636363636362,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048792595160955374,
      "loss": 5.2664,
      "mean_token_accuracy": 0.16343289166688918,
      "num_tokens": 25084318.0,
      "step": 11450
    },
    {
      "entropy": 5.550451278686523,
      "epoch": 1.136859865025804,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048791449822384856,
      "loss": 5.154,
      "mean_token_accuracy": 0.17242386043071747,
      "num_tokens": 25095715.0,
      "step": 11455
    },
    {
      "entropy": 5.570097970962524,
      "epoch": 1.1373560936879714,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004879030395583226,
      "loss": 5.1702,
      "mean_token_accuracy": 0.16954847127199174,
      "num_tokens": 25106163.0,
      "step": 11460
    },
    {
      "entropy": 5.579524564743042,
      "epoch": 1.137852322350139,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004878915756132601,
      "loss": 5.2476,
      "mean_token_accuracy": 0.16438795924186705,
      "num_tokens": 25118745.0,
      "step": 11465
    },
    {
      "entropy": 5.528561401367187,
      "epoch": 1.1383485510123066,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048788010638894523,
      "loss": 5.2274,
      "mean_token_accuracy": 0.15801485627889633,
      "num_tokens": 25130066.0,
      "step": 11470
    },
    {
      "entropy": 5.497963285446167,
      "epoch": 1.138844779674474,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004878686318856624,
      "loss": 5.0723,
      "mean_token_accuracy": 0.1734581857919693,
      "num_tokens": 25139895.0,
      "step": 11475
    },
    {
      "entropy": 5.63120584487915,
      "epoch": 1.1393410083366415,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048785715210369635,
      "loss": 5.3008,
      "mean_token_accuracy": 0.15998752117156984,
      "num_tokens": 25150644.0,
      "step": 11480
    },
    {
      "entropy": 5.549851989746093,
      "epoch": 1.139837236998809,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004878456670433314,
      "loss": 5.2144,
      "mean_token_accuracy": 0.16592725813388826,
      "num_tokens": 25163038.0,
      "step": 11485
    },
    {
      "entropy": 5.490525531768799,
      "epoch": 1.1403334656609765,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004878341767048527,
      "loss": 5.1225,
      "mean_token_accuracy": 0.17211485654115677,
      "num_tokens": 25174154.0,
      "step": 11490
    },
    {
      "entropy": 5.5640708923339846,
      "epoch": 1.1408296943231442,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048782268108854494,
      "loss": 5.2429,
      "mean_token_accuracy": 0.16499737352132798,
      "num_tokens": 25185369.0,
      "step": 11495
    },
    {
      "entropy": 5.511232233047485,
      "epoch": 1.1413259229853117,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004878111801946933,
      "loss": 5.2328,
      "mean_token_accuracy": 0.1616926297545433,
      "num_tokens": 25197875.0,
      "step": 11500
    },
    {
      "entropy": 5.4827024936676025,
      "epoch": 1.1418221516474791,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048779967402358296,
      "loss": 5.1164,
      "mean_token_accuracy": 0.1780473992228508,
      "num_tokens": 25208362.0,
      "step": 11505
    },
    {
      "entropy": 5.589849996566772,
      "epoch": 1.1423183803096466,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004877881625754993,
      "loss": 5.3077,
      "mean_token_accuracy": 0.16069869101047515,
      "num_tokens": 25220441.0,
      "step": 11510
    },
    {
      "entropy": 5.489728593826294,
      "epoch": 1.1428146089718143,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048777664585072765,
      "loss": 5.0851,
      "mean_token_accuracy": 0.16956905275583267,
      "num_tokens": 25231134.0,
      "step": 11515
    },
    {
      "entropy": 5.618770551681519,
      "epoch": 1.1433108376339818,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004877651238495537,
      "loss": 5.3481,
      "mean_token_accuracy": 0.15938334316015243,
      "num_tokens": 25242327.0,
      "step": 11520
    },
    {
      "entropy": 5.551085948944092,
      "epoch": 1.1438070662961493,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048775359657226316,
      "loss": 5.1846,
      "mean_token_accuracy": 0.16684814542531967,
      "num_tokens": 25253571.0,
      "step": 11525
    },
    {
      "entropy": 5.470210313796997,
      "epoch": 1.1443032949583167,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004877420640191418,
      "loss": 5.1238,
      "mean_token_accuracy": 0.17145076692104338,
      "num_tokens": 25264573.0,
      "step": 11530
    },
    {
      "entropy": 5.529512643814087,
      "epoch": 1.1447995236204842,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004877305261904758,
      "loss": 5.2137,
      "mean_token_accuracy": 0.1657499223947525,
      "num_tokens": 25275197.0,
      "step": 11535
    },
    {
      "entropy": 5.516482925415039,
      "epoch": 1.145295752282652,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048771898308655107,
      "loss": 5.097,
      "mean_token_accuracy": 0.17488718926906585,
      "num_tokens": 25286564.0,
      "step": 11540
    },
    {
      "entropy": 5.564264726638794,
      "epoch": 1.1457919809448194,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004877074347076539,
      "loss": 5.1505,
      "mean_token_accuracy": 0.17080543786287308,
      "num_tokens": 25297874.0,
      "step": 11545
    },
    {
      "entropy": 5.543221855163575,
      "epoch": 1.1462882096069869,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048769588105407075,
      "loss": 5.2173,
      "mean_token_accuracy": 0.16473061442375184,
      "num_tokens": 25308047.0,
      "step": 11550
    },
    {
      "entropy": 5.579469966888428,
      "epoch": 1.1467844382691545,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048768432212608807,
      "loss": 5.2563,
      "mean_token_accuracy": 0.1589917406439781,
      "num_tokens": 25318527.0,
      "step": 11555
    },
    {
      "entropy": 5.531277179718018,
      "epoch": 1.147280666931322,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004876727579239925,
      "loss": 5.1707,
      "mean_token_accuracy": 0.17014151066541672,
      "num_tokens": 25328242.0,
      "step": 11560
    },
    {
      "entropy": 5.5312587261199955,
      "epoch": 1.1477768955934895,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004876611884480708,
      "loss": 5.1572,
      "mean_token_accuracy": 0.16783188432455062,
      "num_tokens": 25338214.0,
      "step": 11565
    },
    {
      "entropy": 5.524855613708496,
      "epoch": 1.148273124255657,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048764961369860977,
      "loss": 5.2066,
      "mean_token_accuracy": 0.16303918212652208,
      "num_tokens": 25349762.0,
      "step": 11570
    },
    {
      "entropy": 5.506272411346435,
      "epoch": 1.1487693529178244,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048763803367589663,
      "loss": 5.1949,
      "mean_token_accuracy": 0.16914894729852675,
      "num_tokens": 25361241.0,
      "step": 11575
    },
    {
      "entropy": 5.553636074066162,
      "epoch": 1.1492655815799921,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048762644838021843,
      "loss": 5.2317,
      "mean_token_accuracy": 0.16020854711532592,
      "num_tokens": 25373846.0,
      "step": 11580
    },
    {
      "entropy": 5.527219963073731,
      "epoch": 1.1497618102421596,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048761485781186243,
      "loss": 5.1486,
      "mean_token_accuracy": 0.16828452944755554,
      "num_tokens": 25385331.0,
      "step": 11585
    },
    {
      "entropy": 5.5011152744293215,
      "epoch": 1.150258038904327,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004876032619711162,
      "loss": 5.1743,
      "mean_token_accuracy": 0.17516097724437713,
      "num_tokens": 25395140.0,
      "step": 11590
    },
    {
      "entropy": 5.5760979652404785,
      "epoch": 1.1507542675664946,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048759166085826714,
      "loss": 5.2234,
      "mean_token_accuracy": 0.15750466883182526,
      "num_tokens": 25405455.0,
      "step": 11595
    },
    {
      "entropy": 5.5412132263183596,
      "epoch": 1.1512504962286623,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000487580054473603,
      "loss": 5.2322,
      "mean_token_accuracy": 0.16742897033691406,
      "num_tokens": 25417000.0,
      "step": 11600
    },
    {
      "entropy": 5.440884304046631,
      "epoch": 1.1517467248908297,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048756844281741156,
      "loss": 5.1338,
      "mean_token_accuracy": 0.17535168528556824,
      "num_tokens": 25427808.0,
      "step": 11605
    },
    {
      "entropy": 5.516225719451905,
      "epoch": 1.1522429535529972,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048755682588998085,
      "loss": 5.1628,
      "mean_token_accuracy": 0.1629212662577629,
      "num_tokens": 25438302.0,
      "step": 11610
    },
    {
      "entropy": 5.4552528858184814,
      "epoch": 1.1527391822151647,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004875452036915987,
      "loss": 5.0679,
      "mean_token_accuracy": 0.16725915521383286,
      "num_tokens": 25449335.0,
      "step": 11615
    },
    {
      "entropy": 5.596983766555786,
      "epoch": 1.1532354108773322,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004875335762225536,
      "loss": 5.2561,
      "mean_token_accuracy": 0.16031309962272644,
      "num_tokens": 25461173.0,
      "step": 11620
    },
    {
      "entropy": 5.561198377609253,
      "epoch": 1.1537316395394999,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048752194348313383,
      "loss": 5.1535,
      "mean_token_accuracy": 0.1659163236618042,
      "num_tokens": 25472538.0,
      "step": 11625
    },
    {
      "entropy": 5.49793119430542,
      "epoch": 1.1542278682016673,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004875103054736278,
      "loss": 5.1902,
      "mean_token_accuracy": 0.170088492333889,
      "num_tokens": 25483386.0,
      "step": 11630
    },
    {
      "entropy": 5.547458934783935,
      "epoch": 1.1547240968638348,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004874986621943241,
      "loss": 5.2026,
      "mean_token_accuracy": 0.17032596319913865,
      "num_tokens": 25495505.0,
      "step": 11635
    },
    {
      "entropy": 5.521256303787231,
      "epoch": 1.1552203255260023,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048748701364551147,
      "loss": 5.1882,
      "mean_token_accuracy": 0.16937849819660186,
      "num_tokens": 25506144.0,
      "step": 11640
    },
    {
      "entropy": 5.524990606307983,
      "epoch": 1.15571655418817,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004874753598274787,
      "loss": 5.1835,
      "mean_token_accuracy": 0.1634795621037483,
      "num_tokens": 25516185.0,
      "step": 11645
    },
    {
      "entropy": 5.6168482303619385,
      "epoch": 1.1562127828503375,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004874637007405149,
      "loss": 5.266,
      "mean_token_accuracy": 0.1643349274992943,
      "num_tokens": 25526657.0,
      "step": 11650
    },
    {
      "entropy": 5.4739500999450685,
      "epoch": 1.156709011512505,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048745203638490917,
      "loss": 5.1242,
      "mean_token_accuracy": 0.17248900085687638,
      "num_tokens": 25537101.0,
      "step": 11655
    },
    {
      "entropy": 5.509799146652222,
      "epoch": 1.1572052401746724,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048744036676095067,
      "loss": 5.2268,
      "mean_token_accuracy": 0.15992130041122438,
      "num_tokens": 25547672.0,
      "step": 11660
    },
    {
      "entropy": 5.59275450706482,
      "epoch": 1.15770146883684,
      "grad_norm": 1.0,
      "learning_rate": 0.0004874286918689289,
      "loss": 5.2681,
      "mean_token_accuracy": 0.1656245231628418,
      "num_tokens": 25558408.0,
      "step": 11665
    },
    {
      "entropy": 5.589129781723022,
      "epoch": 1.1581976974990076,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048741701170913326,
      "loss": 5.1705,
      "mean_token_accuracy": 0.1653664618730545,
      "num_tokens": 25569823.0,
      "step": 11670
    },
    {
      "entropy": 5.51324872970581,
      "epoch": 1.158693926161175,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048740532628185353,
      "loss": 5.168,
      "mean_token_accuracy": 0.17340276390314102,
      "num_tokens": 25579705.0,
      "step": 11675
    },
    {
      "entropy": 5.521526765823364,
      "epoch": 1.1591901548233425,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004873936355873793,
      "loss": 5.2393,
      "mean_token_accuracy": 0.1646953508257866,
      "num_tokens": 25590573.0,
      "step": 11680
    },
    {
      "entropy": 5.509283542633057,
      "epoch": 1.1596863834855102,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004873819396260007,
      "loss": 5.1496,
      "mean_token_accuracy": 0.17044238299131392,
      "num_tokens": 25601418.0,
      "step": 11685
    },
    {
      "entropy": 5.5421528816223145,
      "epoch": 1.1601826121476777,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048737023839800754,
      "loss": 5.2243,
      "mean_token_accuracy": 0.164617358148098,
      "num_tokens": 25612409.0,
      "step": 11690
    },
    {
      "entropy": 5.496366214752197,
      "epoch": 1.1606788408098452,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004873585319036901,
      "loss": 5.1403,
      "mean_token_accuracy": 0.17199216336011885,
      "num_tokens": 25623547.0,
      "step": 11695
    },
    {
      "entropy": 5.570702028274536,
      "epoch": 1.1611750694720127,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004873468201433387,
      "loss": 5.319,
      "mean_token_accuracy": 0.160432893037796,
      "num_tokens": 25636537.0,
      "step": 11700
    },
    {
      "entropy": 5.542683935165405,
      "epoch": 1.1616712981341801,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004873351031172437,
      "loss": 5.1433,
      "mean_token_accuracy": 0.1731248676776886,
      "num_tokens": 25646734.0,
      "step": 11705
    },
    {
      "entropy": 5.560605478286743,
      "epoch": 1.1621675267963478,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048732338082569564,
      "loss": 5.2509,
      "mean_token_accuracy": 0.1691226616501808,
      "num_tokens": 25658058.0,
      "step": 11710
    },
    {
      "entropy": 5.492860174179077,
      "epoch": 1.1626637554585153,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004873116532689853,
      "loss": 5.088,
      "mean_token_accuracy": 0.17136604934930802,
      "num_tokens": 25669630.0,
      "step": 11715
    },
    {
      "entropy": 5.450148391723633,
      "epoch": 1.1631599841206828,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004872999204474035,
      "loss": 5.1037,
      "mean_token_accuracy": 0.1707699716091156,
      "num_tokens": 25681928.0,
      "step": 11720
    },
    {
      "entropy": 5.440466070175171,
      "epoch": 1.1636562127828503,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000487288182361241,
      "loss": 5.1108,
      "mean_token_accuracy": 0.1661658838391304,
      "num_tokens": 25690653.0,
      "step": 11725
    },
    {
      "entropy": 5.538868236541748,
      "epoch": 1.164152441445018,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004872764390107891,
      "loss": 5.2163,
      "mean_token_accuracy": 0.16166536659002304,
      "num_tokens": 25701201.0,
      "step": 11730
    },
    {
      "entropy": 5.457179927825928,
      "epoch": 1.1646486701071854,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048726469039633895,
      "loss": 5.133,
      "mean_token_accuracy": 0.17010793089866638,
      "num_tokens": 25710927.0,
      "step": 11735
    },
    {
      "entropy": 5.515346240997315,
      "epoch": 1.165144898769353,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048725293651818183,
      "loss": 5.1783,
      "mean_token_accuracy": 0.16830242574214935,
      "num_tokens": 25722136.0,
      "step": 11740
    },
    {
      "entropy": 5.513708066940308,
      "epoch": 1.1656411274315204,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004872411773766093,
      "loss": 5.0631,
      "mean_token_accuracy": 0.17500446289777755,
      "num_tokens": 25733386.0,
      "step": 11745
    },
    {
      "entropy": 5.467298269271851,
      "epoch": 1.1661373560936878,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048722941297191284,
      "loss": 5.1494,
      "mean_token_accuracy": 0.16997703164815903,
      "num_tokens": 25742801.0,
      "step": 11750
    },
    {
      "entropy": 5.5761877536773685,
      "epoch": 1.1666335847558555,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004872176433043843,
      "loss": 5.2554,
      "mean_token_accuracy": 0.16377057135105133,
      "num_tokens": 25755594.0,
      "step": 11755
    },
    {
      "entropy": 5.559096097946167,
      "epoch": 1.167129813418023,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004872058683743155,
      "loss": 5.187,
      "mean_token_accuracy": 0.1699764460325241,
      "num_tokens": 25766271.0,
      "step": 11760
    },
    {
      "entropy": 5.56765661239624,
      "epoch": 1.1676260420801905,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048719408818199836,
      "loss": 5.2355,
      "mean_token_accuracy": 0.1647241234779358,
      "num_tokens": 25777338.0,
      "step": 11765
    },
    {
      "entropy": 5.510936880111695,
      "epoch": 1.1681222707423582,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048718230272772517,
      "loss": 5.1173,
      "mean_token_accuracy": 0.16859878301620485,
      "num_tokens": 25787968.0,
      "step": 11770
    },
    {
      "entropy": 5.519161939620972,
      "epoch": 1.1686184994045257,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048717051201178804,
      "loss": 5.1988,
      "mean_token_accuracy": 0.17095001190900802,
      "num_tokens": 25798274.0,
      "step": 11775
    },
    {
      "entropy": 5.480053472518921,
      "epoch": 1.1691147280666931,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004871587160344794,
      "loss": 5.158,
      "mean_token_accuracy": 0.16201619505882264,
      "num_tokens": 25809140.0,
      "step": 11780
    },
    {
      "entropy": 5.546340751647949,
      "epoch": 1.1696109567288606,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004871469147960917,
      "loss": 5.1983,
      "mean_token_accuracy": 0.1667727842926979,
      "num_tokens": 25821502.0,
      "step": 11785
    },
    {
      "entropy": 5.529561853408813,
      "epoch": 1.170107185391028,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004871351082969177,
      "loss": 5.0859,
      "mean_token_accuracy": 0.17633658647537231,
      "num_tokens": 25833231.0,
      "step": 11790
    },
    {
      "entropy": 5.568858480453491,
      "epoch": 1.1706034140531958,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004871232965372501,
      "loss": 5.2379,
      "mean_token_accuracy": 0.16485683172941207,
      "num_tokens": 25844867.0,
      "step": 11795
    },
    {
      "entropy": 5.52788896560669,
      "epoch": 1.1710996427153633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004871114795173819,
      "loss": 5.2151,
      "mean_token_accuracy": 0.17525176405906678,
      "num_tokens": 25855959.0,
      "step": 11800
    },
    {
      "entropy": 5.576307678222657,
      "epoch": 1.1715958713775307,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004870996572376061,
      "loss": 5.2943,
      "mean_token_accuracy": 0.16473801285028458,
      "num_tokens": 25866926.0,
      "step": 11805
    },
    {
      "entropy": 5.598384284973145,
      "epoch": 1.1720921000396982,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004870878296982157,
      "loss": 5.228,
      "mean_token_accuracy": 0.16459791511297225,
      "num_tokens": 25877699.0,
      "step": 11810
    },
    {
      "entropy": 5.592456245422364,
      "epoch": 1.172588328701866,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004870759968995042,
      "loss": 5.174,
      "mean_token_accuracy": 0.17203024327754973,
      "num_tokens": 25888640.0,
      "step": 11815
    },
    {
      "entropy": 5.517426633834839,
      "epoch": 1.1730845573640334,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048706415884176493,
      "loss": 5.173,
      "mean_token_accuracy": 0.1669470727443695,
      "num_tokens": 25900651.0,
      "step": 11820
    },
    {
      "entropy": 5.508384704589844,
      "epoch": 1.1735807860262009,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004870523155252915,
      "loss": 5.1593,
      "mean_token_accuracy": 0.16976964324712754,
      "num_tokens": 25911775.0,
      "step": 11825
    },
    {
      "entropy": 5.508563232421875,
      "epoch": 1.1740770146883683,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004870404669503776,
      "loss": 5.1163,
      "mean_token_accuracy": 0.16939260065555573,
      "num_tokens": 25922126.0,
      "step": 11830
    },
    {
      "entropy": 5.565032005310059,
      "epoch": 1.1745732433505358,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000487028613117317,
      "loss": 5.1804,
      "mean_token_accuracy": 0.1685253396630287,
      "num_tokens": 25932945.0,
      "step": 11835
    },
    {
      "entropy": 5.596076774597168,
      "epoch": 1.1750694720127035,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004870167540264037,
      "loss": 5.2499,
      "mean_token_accuracy": 0.15927001237869262,
      "num_tokens": 25944277.0,
      "step": 11840
    },
    {
      "entropy": 5.453707075119018,
      "epoch": 1.175565700674871,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004870048896779317,
      "loss": 5.1124,
      "mean_token_accuracy": 0.17226413041353225,
      "num_tokens": 25956838.0,
      "step": 11845
    },
    {
      "entropy": 5.478300619125366,
      "epoch": 1.1760619293370385,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048699302007219533,
      "loss": 5.1597,
      "mean_token_accuracy": 0.16248856782913207,
      "num_tokens": 25966639.0,
      "step": 11850
    },
    {
      "entropy": 5.521308898925781,
      "epoch": 1.1765581579992062,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004869811452094889,
      "loss": 5.2009,
      "mean_token_accuracy": 0.173517669737339,
      "num_tokens": 25978209.0,
      "step": 11855
    },
    {
      "entropy": 5.498318338394165,
      "epoch": 1.1770543866613736,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048696926509010673,
      "loss": 5.1334,
      "mean_token_accuracy": 0.1738154858350754,
      "num_tokens": 25989306.0,
      "step": 11860
    },
    {
      "entropy": 5.537467432022095,
      "epoch": 1.177550615323541,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004869573797143437,
      "loss": 5.2327,
      "mean_token_accuracy": 0.16106203198432922,
      "num_tokens": 26000235.0,
      "step": 11865
    },
    {
      "entropy": 5.511281061172485,
      "epoch": 1.1780468439857086,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004869454890824942,
      "loss": 5.1268,
      "mean_token_accuracy": 0.17023174166679383,
      "num_tokens": 26010616.0,
      "step": 11870
    },
    {
      "entropy": 5.511196327209473,
      "epoch": 1.178543072647876,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00048693359319485337,
      "loss": 5.2079,
      "mean_token_accuracy": 0.16644639074802398,
      "num_tokens": 26021303.0,
      "step": 11875
    },
    {
      "entropy": 5.527012968063355,
      "epoch": 1.1790393013100438,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004869216920517161,
      "loss": 5.1298,
      "mean_token_accuracy": 0.16764458566904067,
      "num_tokens": 26032386.0,
      "step": 11880
    },
    {
      "entropy": 5.454592275619507,
      "epoch": 1.1795355299722112,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048690978565337753,
      "loss": 5.0962,
      "mean_token_accuracy": 0.1689310885965824,
      "num_tokens": 26042889.0,
      "step": 11885
    },
    {
      "entropy": 5.479773139953613,
      "epoch": 1.1800317586343787,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048689787400013287,
      "loss": 5.2047,
      "mean_token_accuracy": 0.16476578414440154,
      "num_tokens": 26054299.0,
      "step": 11890
    },
    {
      "entropy": 5.61276683807373,
      "epoch": 1.1805279872965462,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004868859570922775,
      "loss": 5.2453,
      "mean_token_accuracy": 0.16201778948307038,
      "num_tokens": 26064587.0,
      "step": 11895
    },
    {
      "entropy": 5.580943870544433,
      "epoch": 1.1810242159587139,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004868740349301071,
      "loss": 5.1712,
      "mean_token_accuracy": 0.17701984643936158,
      "num_tokens": 26076495.0,
      "step": 11900
    },
    {
      "entropy": 5.543245077133179,
      "epoch": 1.1815204446208813,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000486862107513917,
      "loss": 5.1831,
      "mean_token_accuracy": 0.16747597455978394,
      "num_tokens": 26086496.0,
      "step": 11905
    },
    {
      "entropy": 5.488832187652588,
      "epoch": 1.1820166732830488,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004868501748440033,
      "loss": 5.157,
      "mean_token_accuracy": 0.16652124971151352,
      "num_tokens": 26097263.0,
      "step": 11910
    },
    {
      "entropy": 5.536653661727906,
      "epoch": 1.1825129019452163,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004868382369206617,
      "loss": 5.161,
      "mean_token_accuracy": 0.168246990442276,
      "num_tokens": 26108264.0,
      "step": 11915
    },
    {
      "entropy": 5.532262706756592,
      "epoch": 1.1830091306073838,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004868262937441883,
      "loss": 5.2233,
      "mean_token_accuracy": 0.17646037936210632,
      "num_tokens": 26119730.0,
      "step": 11920
    },
    {
      "entropy": 5.55259370803833,
      "epoch": 1.1835053592695515,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048681434531487925,
      "loss": 5.1289,
      "mean_token_accuracy": 0.16772469878196716,
      "num_tokens": 26131055.0,
      "step": 11925
    },
    {
      "entropy": 5.4555566787719725,
      "epoch": 1.184001587931719,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048680239163303085,
      "loss": 5.121,
      "mean_token_accuracy": 0.17310939133167266,
      "num_tokens": 26141577.0,
      "step": 11930
    },
    {
      "entropy": 5.5206818103790285,
      "epoch": 1.1844978165938864,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004867904326989396,
      "loss": 5.1716,
      "mean_token_accuracy": 0.16741517931222916,
      "num_tokens": 26152733.0,
      "step": 11935
    },
    {
      "entropy": 5.569200706481934,
      "epoch": 1.184994045256054,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004867784685129019,
      "loss": 5.2112,
      "mean_token_accuracy": 0.16788169592618943,
      "num_tokens": 26164763.0,
      "step": 11940
    },
    {
      "entropy": 5.486721515655518,
      "epoch": 1.1854902739182216,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048676649907521455,
      "loss": 5.0908,
      "mean_token_accuracy": 0.17138195186853408,
      "num_tokens": 26174391.0,
      "step": 11945
    },
    {
      "entropy": 5.417409515380859,
      "epoch": 1.185986502580389,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048675452438617434,
      "loss": 5.0737,
      "mean_token_accuracy": 0.17694325000047684,
      "num_tokens": 26184772.0,
      "step": 11950
    },
    {
      "entropy": 5.583131980895996,
      "epoch": 1.1864827312425565,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004867425444460782,
      "loss": 5.2208,
      "mean_token_accuracy": 0.15865879356861115,
      "num_tokens": 26195411.0,
      "step": 11955
    },
    {
      "entropy": 5.488093328475952,
      "epoch": 1.186978959904724,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004867305592552233,
      "loss": 5.1463,
      "mean_token_accuracy": 0.16958065927028657,
      "num_tokens": 26206719.0,
      "step": 11960
    },
    {
      "entropy": 5.47322154045105,
      "epoch": 1.1874751885668917,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004867185688139067,
      "loss": 5.0898,
      "mean_token_accuracy": 0.17300083339214326,
      "num_tokens": 26217469.0,
      "step": 11965
    },
    {
      "entropy": 5.454719018936157,
      "epoch": 1.1879714172290592,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004867065731224258,
      "loss": 5.1056,
      "mean_token_accuracy": 0.16581848561763762,
      "num_tokens": 26228556.0,
      "step": 11970
    },
    {
      "entropy": 5.533361625671387,
      "epoch": 1.1884676458912267,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004866945721810781,
      "loss": 5.2115,
      "mean_token_accuracy": 0.16659659743309022,
      "num_tokens": 26239891.0,
      "step": 11975
    },
    {
      "entropy": 5.521927070617676,
      "epoch": 1.1889638745533941,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004866825659901612,
      "loss": 5.1733,
      "mean_token_accuracy": 0.1729780450463295,
      "num_tokens": 26251793.0,
      "step": 11980
    },
    {
      "entropy": 5.426382732391358,
      "epoch": 1.1894601032155618,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048667055454997276,
      "loss": 5.0653,
      "mean_token_accuracy": 0.17323132008314132,
      "num_tokens": 26262367.0,
      "step": 11985
    },
    {
      "entropy": 5.483523797988892,
      "epoch": 1.1899563318777293,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048665853786081063,
      "loss": 5.1624,
      "mean_token_accuracy": 0.167740099132061,
      "num_tokens": 26272681.0,
      "step": 11990
    },
    {
      "entropy": 5.540319919586182,
      "epoch": 1.1904525605398968,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048664651592297294,
      "loss": 5.2232,
      "mean_token_accuracy": 0.16316691040992737,
      "num_tokens": 26281194.0,
      "step": 11995
    },
    {
      "entropy": 5.543879842758178,
      "epoch": 1.1909487892020643,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048663448873675766,
      "loss": 5.1236,
      "mean_token_accuracy": 0.1654999375343323,
      "num_tokens": 26292682.0,
      "step": 12000
    },
    {
      "epoch": 1.1909487892020643,
      "eval_entropy": 5.388492889997936,
      "eval_loss": 5.299468517303467,
      "eval_mean_token_accuracy": 0.1675439624137509,
      "eval_num_tokens": 26292682.0,
      "eval_runtime": 20.8063,
      "eval_samples_per_second": 1618.211,
      "eval_steps_per_second": 202.294,
      "step": 12000
    },
    {
      "entropy": 5.559010601043701,
      "epoch": 1.1914450178642317,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004866224563024631,
      "loss": 5.2223,
      "mean_token_accuracy": 0.16385785341262818,
      "num_tokens": 26305500.0,
      "step": 12005
    },
    {
      "entropy": 5.515285015106201,
      "epoch": 1.1919412465263994,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004866104186203877,
      "loss": 5.1131,
      "mean_token_accuracy": 0.1688894212245941,
      "num_tokens": 26317207.0,
      "step": 12010
    },
    {
      "entropy": 5.517972230911255,
      "epoch": 1.192437475188567,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048659837569082977,
      "loss": 5.1217,
      "mean_token_accuracy": 0.1721375048160553,
      "num_tokens": 26328728.0,
      "step": 12015
    },
    {
      "entropy": 5.580038928985596,
      "epoch": 1.1929337038507344,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004865863275140882,
      "loss": 5.1815,
      "mean_token_accuracy": 0.1676922097802162,
      "num_tokens": 26338622.0,
      "step": 12020
    },
    {
      "entropy": 5.538651657104492,
      "epoch": 1.1934299325129019,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004865742740904615,
      "loss": 5.1573,
      "mean_token_accuracy": 0.1691615954041481,
      "num_tokens": 26348807.0,
      "step": 12025
    },
    {
      "entropy": 5.446991109848023,
      "epoch": 1.1939261611750696,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004865622154202488,
      "loss": 5.1225,
      "mean_token_accuracy": 0.17199401259422303,
      "num_tokens": 26358481.0,
      "step": 12030
    },
    {
      "entropy": 5.521640825271606,
      "epoch": 1.194422389837237,
      "grad_norm": 0.875,
      "learning_rate": 0.00048655015150374904,
      "loss": 5.1968,
      "mean_token_accuracy": 0.16276587247848512,
      "num_tokens": 26370519.0,
      "step": 12035
    },
    {
      "entropy": 5.566204071044922,
      "epoch": 1.1949186184994045,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004865380823412613,
      "loss": 5.196,
      "mean_token_accuracy": 0.17306286990642547,
      "num_tokens": 26380988.0,
      "step": 12040
    },
    {
      "entropy": 5.499672985076904,
      "epoch": 1.195414847161572,
      "grad_norm": 0.96875,
      "learning_rate": 0.000486526007933085,
      "loss": 5.0886,
      "mean_token_accuracy": 0.17434173077344894,
      "num_tokens": 26390308.0,
      "step": 12045
    },
    {
      "entropy": 5.5383542537689205,
      "epoch": 1.1959110758237395,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048651392827951944,
      "loss": 5.2133,
      "mean_token_accuracy": 0.1591360315680504,
      "num_tokens": 26401145.0,
      "step": 12050
    },
    {
      "entropy": 5.5039387226104735,
      "epoch": 1.1964073044859072,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048650184338086436,
      "loss": 5.1461,
      "mean_token_accuracy": 0.16698650270700455,
      "num_tokens": 26411930.0,
      "step": 12055
    },
    {
      "entropy": 5.519377422332764,
      "epoch": 1.1969035331480746,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048648975323741916,
      "loss": 5.0848,
      "mean_token_accuracy": 0.16916512101888656,
      "num_tokens": 26422320.0,
      "step": 12060
    },
    {
      "entropy": 5.638296699523925,
      "epoch": 1.197399761810242,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048647765784948395,
      "loss": 5.2167,
      "mean_token_accuracy": 0.16180272549390792,
      "num_tokens": 26433760.0,
      "step": 12065
    },
    {
      "entropy": 5.5516104221344,
      "epoch": 1.1978959904724098,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004864655572173584,
      "loss": 5.131,
      "mean_token_accuracy": 0.1768320232629776,
      "num_tokens": 26444256.0,
      "step": 12070
    },
    {
      "entropy": 5.507214641571045,
      "epoch": 1.1983922191345773,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048645345134134274,
      "loss": 5.2217,
      "mean_token_accuracy": 0.16871612817049025,
      "num_tokens": 26456433.0,
      "step": 12075
    },
    {
      "entropy": 5.551804685592652,
      "epoch": 1.1988884477967447,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048644134022173707,
      "loss": 5.2021,
      "mean_token_accuracy": 0.16770168393850327,
      "num_tokens": 26466909.0,
      "step": 12080
    },
    {
      "entropy": 5.5663196563720705,
      "epoch": 1.1993846764589122,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004864292238588419,
      "loss": 5.1945,
      "mean_token_accuracy": 0.1715540662407875,
      "num_tokens": 26478785.0,
      "step": 12085
    },
    {
      "entropy": 5.50177788734436,
      "epoch": 1.1998809051210797,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048641710225295746,
      "loss": 5.2012,
      "mean_token_accuracy": 0.16975893527269365,
      "num_tokens": 26489596.0,
      "step": 12090
    },
    {
      "entropy": 5.4807556629180905,
      "epoch": 1.2003771337832474,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048640497540438455,
      "loss": 5.1388,
      "mean_token_accuracy": 0.17157377004623414,
      "num_tokens": 26500376.0,
      "step": 12095
    },
    {
      "entropy": 5.503072261810303,
      "epoch": 1.2008733624454149,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004863928433134237,
      "loss": 5.0882,
      "mean_token_accuracy": 0.16729052662849425,
      "num_tokens": 26510339.0,
      "step": 12100
    },
    {
      "entropy": 5.528472137451172,
      "epoch": 1.2013695911075823,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048638070598037585,
      "loss": 5.1794,
      "mean_token_accuracy": 0.16958650946617126,
      "num_tokens": 26521359.0,
      "step": 12105
    },
    {
      "entropy": 5.578713893890381,
      "epoch": 1.2018658197697498,
      "grad_norm": 1.03125,
      "learning_rate": 0.000486368563405542,
      "loss": 5.2437,
      "mean_token_accuracy": 0.16707834005355834,
      "num_tokens": 26532716.0,
      "step": 12110
    },
    {
      "entropy": 5.496238422393799,
      "epoch": 1.2023620484319175,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004863564155892232,
      "loss": 5.0969,
      "mean_token_accuracy": 0.1705687537789345,
      "num_tokens": 26542624.0,
      "step": 12115
    },
    {
      "entropy": 5.566644954681396,
      "epoch": 1.202858277094085,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004863442625317208,
      "loss": 5.2012,
      "mean_token_accuracy": 0.17068732231855394,
      "num_tokens": 26552907.0,
      "step": 12120
    },
    {
      "entropy": 5.484821224212647,
      "epoch": 1.2033545057562525,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048633210423333607,
      "loss": 5.1571,
      "mean_token_accuracy": 0.1678165689110756,
      "num_tokens": 26564871.0,
      "step": 12125
    },
    {
      "entropy": 5.482084131240844,
      "epoch": 1.20385073441842,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004863199406943705,
      "loss": 5.1392,
      "mean_token_accuracy": 0.17085685431957245,
      "num_tokens": 26576254.0,
      "step": 12130
    },
    {
      "entropy": 5.509066438674926,
      "epoch": 1.2043469630805874,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048630777191512584,
      "loss": 5.1693,
      "mean_token_accuracy": 0.17062569707632064,
      "num_tokens": 26587518.0,
      "step": 12135
    },
    {
      "entropy": 5.483522033691406,
      "epoch": 1.2048431917427551,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048629559789590366,
      "loss": 5.0852,
      "mean_token_accuracy": 0.1703111633658409,
      "num_tokens": 26597485.0,
      "step": 12140
    },
    {
      "entropy": 5.469754028320312,
      "epoch": 1.2053394204049226,
      "grad_norm": 0.984375,
      "learning_rate": 0.000486283418637006,
      "loss": 5.1367,
      "mean_token_accuracy": 0.1717156320810318,
      "num_tokens": 26607696.0,
      "step": 12145
    },
    {
      "entropy": 5.552721881866455,
      "epoch": 1.20583564906709,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004862712341387348,
      "loss": 5.1612,
      "mean_token_accuracy": 0.1707509994506836,
      "num_tokens": 26618974.0,
      "step": 12150
    },
    {
      "entropy": 5.576195907592774,
      "epoch": 1.2063318777292578,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004862590444013923,
      "loss": 5.2022,
      "mean_token_accuracy": 0.1651910662651062,
      "num_tokens": 26629949.0,
      "step": 12155
    },
    {
      "entropy": 5.54221248626709,
      "epoch": 1.2068281063914252,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004862468494252806,
      "loss": 5.1396,
      "mean_token_accuracy": 0.16881280839443208,
      "num_tokens": 26641460.0,
      "step": 12160
    },
    {
      "entropy": 5.530866956710815,
      "epoch": 1.2073243350535927,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004862346492107024,
      "loss": 5.2338,
      "mean_token_accuracy": 0.16301158666610718,
      "num_tokens": 26652553.0,
      "step": 12165
    },
    {
      "entropy": 5.480940151214599,
      "epoch": 1.2078205637157602,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048622244375796,
      "loss": 5.1511,
      "mean_token_accuracy": 0.16281547099351884,
      "num_tokens": 26662479.0,
      "step": 12170
    },
    {
      "entropy": 5.560932159423828,
      "epoch": 1.2083167923779277,
      "grad_norm": 0.875,
      "learning_rate": 0.0004862102330673561,
      "loss": 5.2016,
      "mean_token_accuracy": 0.16211413890123366,
      "num_tokens": 26673326.0,
      "step": 12175
    },
    {
      "entropy": 5.589660501480102,
      "epoch": 1.2088130210400954,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048619801713919357,
      "loss": 5.258,
      "mean_token_accuracy": 0.16120460033416747,
      "num_tokens": 26684189.0,
      "step": 12180
    },
    {
      "entropy": 5.592612171173096,
      "epoch": 1.2093092497022628,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004861857959737753,
      "loss": 5.2566,
      "mean_token_accuracy": 0.1649405136704445,
      "num_tokens": 26695399.0,
      "step": 12185
    },
    {
      "entropy": 5.570272493362427,
      "epoch": 1.2098054783644303,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004861735695714044,
      "loss": 5.1903,
      "mean_token_accuracy": 0.16377503871917726,
      "num_tokens": 26705993.0,
      "step": 12190
    },
    {
      "entropy": 5.542041778564453,
      "epoch": 1.2103017070265978,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048616133793238394,
      "loss": 5.1604,
      "mean_token_accuracy": 0.16853890866041182,
      "num_tokens": 26716120.0,
      "step": 12195
    },
    {
      "entropy": 5.574769496917725,
      "epoch": 1.2107979356887655,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048614910105701737,
      "loss": 5.2868,
      "mean_token_accuracy": 0.15985732674598693,
      "num_tokens": 26727736.0,
      "step": 12200
    },
    {
      "entropy": 5.582148265838623,
      "epoch": 1.211294164350933,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004861368589456081,
      "loss": 5.1344,
      "mean_token_accuracy": 0.17238374352455138,
      "num_tokens": 26737825.0,
      "step": 12205
    },
    {
      "entropy": 5.6176838874816895,
      "epoch": 1.2117903930131004,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004861246115984596,
      "loss": 5.2256,
      "mean_token_accuracy": 0.16376420259475707,
      "num_tokens": 26748035.0,
      "step": 12210
    },
    {
      "entropy": 5.570216703414917,
      "epoch": 1.212286621675268,
      "grad_norm": 0.8125,
      "learning_rate": 0.00048611235901587575,
      "loss": 5.1437,
      "mean_token_accuracy": 0.17248266339302062,
      "num_tokens": 26759899.0,
      "step": 12215
    },
    {
      "entropy": 5.483235931396484,
      "epoch": 1.2127828503374354,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004861001011981602,
      "loss": 5.0948,
      "mean_token_accuracy": 0.174762000143528,
      "num_tokens": 26771562.0,
      "step": 12220
    },
    {
      "entropy": 5.572576427459717,
      "epoch": 1.213279078999603,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004860878381456171,
      "loss": 5.2247,
      "mean_token_accuracy": 0.1695061057806015,
      "num_tokens": 26783002.0,
      "step": 12225
    },
    {
      "entropy": 5.534043502807617,
      "epoch": 1.2137753076617706,
      "grad_norm": 1.0,
      "learning_rate": 0.00048607556985855047,
      "loss": 5.2025,
      "mean_token_accuracy": 0.16991137713193893,
      "num_tokens": 26794136.0,
      "step": 12230
    },
    {
      "entropy": 5.45085597038269,
      "epoch": 1.214271536323938,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048606329633726447,
      "loss": 5.1786,
      "mean_token_accuracy": 0.16698249131441117,
      "num_tokens": 26804776.0,
      "step": 12235
    },
    {
      "entropy": 5.497370624542237,
      "epoch": 1.2147677649861055,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004860510175820636,
      "loss": 5.106,
      "mean_token_accuracy": 0.17415401041507722,
      "num_tokens": 26816696.0,
      "step": 12240
    },
    {
      "entropy": 5.462891387939453,
      "epoch": 1.2152639936482732,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004860387335932522,
      "loss": 5.1486,
      "mean_token_accuracy": 0.1721884787082672,
      "num_tokens": 26827789.0,
      "step": 12245
    },
    {
      "entropy": 5.540329074859619,
      "epoch": 1.2157602223104407,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004860264443711351,
      "loss": 5.1082,
      "mean_token_accuracy": 0.17041248679161072,
      "num_tokens": 26838763.0,
      "step": 12250
    },
    {
      "entropy": 5.461474275588989,
      "epoch": 1.2162564509726081,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004860141499160167,
      "loss": 5.1394,
      "mean_token_accuracy": 0.16416517943143843,
      "num_tokens": 26849304.0,
      "step": 12255
    },
    {
      "entropy": 5.502264976501465,
      "epoch": 1.2167526796347756,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004860018502282022,
      "loss": 5.2206,
      "mean_token_accuracy": 0.1691025137901306,
      "num_tokens": 26858777.0,
      "step": 12260
    },
    {
      "entropy": 5.557392120361328,
      "epoch": 1.2172489082969433,
      "grad_norm": 0.85546875,
      "learning_rate": 0.00048598954530799644,
      "loss": 5.1737,
      "mean_token_accuracy": 0.1635236620903015,
      "num_tokens": 26870286.0,
      "step": 12265
    },
    {
      "entropy": 5.545029783248902,
      "epoch": 1.2177451369591108,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048597723515570465,
      "loss": 5.1466,
      "mean_token_accuracy": 0.17426633685827256,
      "num_tokens": 26880485.0,
      "step": 12270
    },
    {
      "entropy": 5.437982892990112,
      "epoch": 1.2182413656212783,
      "grad_norm": 0.8125,
      "learning_rate": 0.000485964919771632,
      "loss": 5.0658,
      "mean_token_accuracy": 0.17580669820308686,
      "num_tokens": 26893556.0,
      "step": 12275
    },
    {
      "entropy": 5.494607257843017,
      "epoch": 1.2187375942834457,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048595259915608396,
      "loss": 5.1416,
      "mean_token_accuracy": 0.16553050130605698,
      "num_tokens": 26905699.0,
      "step": 12280
    },
    {
      "entropy": 5.532801342010498,
      "epoch": 1.2192338229456134,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048594027330936594,
      "loss": 5.1605,
      "mean_token_accuracy": 0.16534959375858307,
      "num_tokens": 26917188.0,
      "step": 12285
    },
    {
      "entropy": 5.546386432647705,
      "epoch": 1.219730051607781,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048592794223178374,
      "loss": 5.1587,
      "mean_token_accuracy": 0.16714199930429458,
      "num_tokens": 26927905.0,
      "step": 12290
    },
    {
      "entropy": 5.456134510040283,
      "epoch": 1.2202262802699484,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048591560592364314,
      "loss": 5.1271,
      "mean_token_accuracy": 0.17412807792425156,
      "num_tokens": 26938665.0,
      "step": 12295
    },
    {
      "entropy": 5.508511304855347,
      "epoch": 1.2207225089321159,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004859032643852499,
      "loss": 5.2274,
      "mean_token_accuracy": 0.16469652503728865,
      "num_tokens": 26950516.0,
      "step": 12300
    },
    {
      "entropy": 5.593306541442871,
      "epoch": 1.2212187375942833,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004858909176169102,
      "loss": 5.2159,
      "mean_token_accuracy": 0.17017593383789062,
      "num_tokens": 26960018.0,
      "step": 12305
    },
    {
      "entropy": 5.524492979049683,
      "epoch": 1.221714966256451,
      "grad_norm": 1.0,
      "learning_rate": 0.00048587856561893014,
      "loss": 5.1814,
      "mean_token_accuracy": 0.16484884470701217,
      "num_tokens": 26969825.0,
      "step": 12310
    },
    {
      "entropy": 5.563662004470825,
      "epoch": 1.2222111949186185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004858662083916161,
      "loss": 5.2684,
      "mean_token_accuracy": 0.16246035248041152,
      "num_tokens": 26979873.0,
      "step": 12315
    },
    {
      "entropy": 5.585719299316406,
      "epoch": 1.222707423580786,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004858538459352744,
      "loss": 5.25,
      "mean_token_accuracy": 0.1645125925540924,
      "num_tokens": 26990577.0,
      "step": 12320
    },
    {
      "entropy": 5.519025754928589,
      "epoch": 1.2232036522429535,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004858414782502117,
      "loss": 5.0913,
      "mean_token_accuracy": 0.1726545885205269,
      "num_tokens": 27001598.0,
      "step": 12325
    },
    {
      "entropy": 5.451758193969726,
      "epoch": 1.2236998809051212,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048582910533673474,
      "loss": 5.0994,
      "mean_token_accuracy": 0.17842347621917726,
      "num_tokens": 27012535.0,
      "step": 12330
    },
    {
      "entropy": 5.509342813491822,
      "epoch": 1.2241961095672886,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004858167271951502,
      "loss": 5.1768,
      "mean_token_accuracy": 0.16546874791383742,
      "num_tokens": 27021867.0,
      "step": 12335
    },
    {
      "entropy": 5.454013967514038,
      "epoch": 1.2246923382294561,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004858043438257651,
      "loss": 5.1572,
      "mean_token_accuracy": 0.1679985836148262,
      "num_tokens": 27032540.0,
      "step": 12340
    },
    {
      "entropy": 5.428150987625122,
      "epoch": 1.2251885668916236,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004857919552288865,
      "loss": 5.1154,
      "mean_token_accuracy": 0.1721821054816246,
      "num_tokens": 27044008.0,
      "step": 12345
    },
    {
      "entropy": 5.546235132217407,
      "epoch": 1.225684795553791,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004857795614048217,
      "loss": 5.1418,
      "mean_token_accuracy": 0.17591281235218048,
      "num_tokens": 27053646.0,
      "step": 12350
    },
    {
      "entropy": 5.612762546539306,
      "epoch": 1.2261810242159588,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048576716235387786,
      "loss": 5.2688,
      "mean_token_accuracy": 0.16229404360055924,
      "num_tokens": 27065834.0,
      "step": 12355
    },
    {
      "entropy": 5.516036462783814,
      "epoch": 1.2266772528781262,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048575475807636267,
      "loss": 5.2008,
      "mean_token_accuracy": 0.1633292019367218,
      "num_tokens": 27077544.0,
      "step": 12360
    },
    {
      "entropy": 5.559359359741211,
      "epoch": 1.2271734815402937,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048574234857258356,
      "loss": 5.2486,
      "mean_token_accuracy": 0.1663970723748207,
      "num_tokens": 27088294.0,
      "step": 12365
    },
    {
      "entropy": 5.554310035705567,
      "epoch": 1.2276697102024614,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048572993384284833,
      "loss": 5.1804,
      "mean_token_accuracy": 0.16529619842767715,
      "num_tokens": 27098848.0,
      "step": 12370
    },
    {
      "entropy": 5.4490556716918945,
      "epoch": 1.2281659388646289,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004857175138874648,
      "loss": 5.1498,
      "mean_token_accuracy": 0.17082599699497222,
      "num_tokens": 27112319.0,
      "step": 12375
    },
    {
      "entropy": 5.548217105865478,
      "epoch": 1.2286621675267964,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048570508870674097,
      "loss": 5.2044,
      "mean_token_accuracy": 0.17085884213447572,
      "num_tokens": 27121840.0,
      "step": 12380
    },
    {
      "entropy": 5.558666038513183,
      "epoch": 1.2291583961889638,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048569265830098493,
      "loss": 5.2095,
      "mean_token_accuracy": 0.16960825473070146,
      "num_tokens": 27133619.0,
      "step": 12385
    },
    {
      "entropy": 5.523226165771485,
      "epoch": 1.2296546248511313,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000485680222670505,
      "loss": 5.1867,
      "mean_token_accuracy": 0.1673790395259857,
      "num_tokens": 27144161.0,
      "step": 12390
    },
    {
      "entropy": 5.499545097351074,
      "epoch": 1.230150853513299,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004856677818156096,
      "loss": 5.1088,
      "mean_token_accuracy": 0.17127839028835296,
      "num_tokens": 27154906.0,
      "step": 12395
    },
    {
      "entropy": 5.50167932510376,
      "epoch": 1.2306470821754665,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048565533573660707,
      "loss": 5.1376,
      "mean_token_accuracy": 0.1709315612912178,
      "num_tokens": 27165746.0,
      "step": 12400
    },
    {
      "entropy": 5.4959306716918945,
      "epoch": 1.231143310837634,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004856428844338062,
      "loss": 5.1221,
      "mean_token_accuracy": 0.1723973944783211,
      "num_tokens": 27176108.0,
      "step": 12405
    },
    {
      "entropy": 5.496760559082031,
      "epoch": 1.2316395394998014,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004856304279075156,
      "loss": 5.2017,
      "mean_token_accuracy": 0.1688024580478668,
      "num_tokens": 27186721.0,
      "step": 12410
    },
    {
      "entropy": 5.500027847290039,
      "epoch": 1.2321357681619691,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004856179661580444,
      "loss": 5.0749,
      "mean_token_accuracy": 0.17624875605106355,
      "num_tokens": 27196120.0,
      "step": 12415
    },
    {
      "entropy": 5.55590295791626,
      "epoch": 1.2326319968241366,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004856054991857013,
      "loss": 5.2036,
      "mean_token_accuracy": 0.1612488016486168,
      "num_tokens": 27207431.0,
      "step": 12420
    },
    {
      "entropy": 5.4861139297485355,
      "epoch": 1.233128225486304,
      "grad_norm": 0.79296875,
      "learning_rate": 0.00048559302699079576,
      "loss": 5.0917,
      "mean_token_accuracy": 0.1792071908712387,
      "num_tokens": 27218489.0,
      "step": 12425
    },
    {
      "entropy": 5.582187223434448,
      "epoch": 1.2336244541484715,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004855805495736369,
      "loss": 5.2163,
      "mean_token_accuracy": 0.1659310668706894,
      "num_tokens": 27230849.0,
      "step": 12430
    },
    {
      "entropy": 5.551375389099121,
      "epoch": 1.234120682810639,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004855680669345342,
      "loss": 5.1717,
      "mean_token_accuracy": 0.16809328496456147,
      "num_tokens": 27242838.0,
      "step": 12435
    },
    {
      "entropy": 5.552640104293824,
      "epoch": 1.2346169114728067,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00048555557907379714,
      "loss": 5.1923,
      "mean_token_accuracy": 0.17570284008979797,
      "num_tokens": 27253945.0,
      "step": 12440
    },
    {
      "entropy": 5.518435430526734,
      "epoch": 1.2351131401349742,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048554308599173544,
      "loss": 5.2638,
      "mean_token_accuracy": 0.16034649312496185,
      "num_tokens": 27263999.0,
      "step": 12445
    },
    {
      "entropy": 5.5560914993286135,
      "epoch": 1.2356093687971417,
      "grad_norm": 1.0,
      "learning_rate": 0.0004855305876886589,
      "loss": 5.0993,
      "mean_token_accuracy": 0.175088270008564,
      "num_tokens": 27274677.0,
      "step": 12450
    },
    {
      "entropy": 5.53433256149292,
      "epoch": 1.2361055974593094,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00048551808416487743,
      "loss": 5.1836,
      "mean_token_accuracy": 0.16513915061950685,
      "num_tokens": 27286512.0,
      "step": 12455
    },
    {
      "entropy": 5.4974696159362795,
      "epoch": 1.2366018261214768,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004855055754207011,
      "loss": 5.1836,
      "mean_token_accuracy": 0.17457702606916428,
      "num_tokens": 27297620.0,
      "step": 12460
    },
    {
      "entropy": 5.564794731140137,
      "epoch": 1.2370980547836443,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004854930614564401,
      "loss": 5.1921,
      "mean_token_accuracy": 0.16229198276996612,
      "num_tokens": 27309963.0,
      "step": 12465
    },
    {
      "entropy": 5.628265762329102,
      "epoch": 1.2375942834458118,
      "grad_norm": 1.0,
      "learning_rate": 0.0004854805422724047,
      "loss": 5.2779,
      "mean_token_accuracy": 0.16460468918085097,
      "num_tokens": 27320740.0,
      "step": 12470
    },
    {
      "entropy": 5.462673139572144,
      "epoch": 1.2380905121079793,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004854680178689055,
      "loss": 5.0968,
      "mean_token_accuracy": 0.17474846988916398,
      "num_tokens": 27331543.0,
      "step": 12475
    },
    {
      "entropy": 5.442281913757324,
      "epoch": 1.238586740770147,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048545548824625287,
      "loss": 5.0829,
      "mean_token_accuracy": 0.17312445342540742,
      "num_tokens": 27343222.0,
      "step": 12480
    },
    {
      "entropy": 5.532667112350464,
      "epoch": 1.2390829694323144,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004854429534047577,
      "loss": 5.1463,
      "mean_token_accuracy": 0.16827509850263594,
      "num_tokens": 27354694.0,
      "step": 12485
    },
    {
      "entropy": 5.481135368347168,
      "epoch": 1.239579198094482,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004854304133447307,
      "loss": 5.1338,
      "mean_token_accuracy": 0.1677487835288048,
      "num_tokens": 27365462.0,
      "step": 12490
    },
    {
      "entropy": 5.442475605010986,
      "epoch": 1.2400754267566494,
      "grad_norm": 0.875,
      "learning_rate": 0.00048541786806648293,
      "loss": 5.0657,
      "mean_token_accuracy": 0.1727943554520607,
      "num_tokens": 27377548.0,
      "step": 12495
    },
    {
      "entropy": 5.555824899673462,
      "epoch": 1.240571655418817,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004854053175703254,
      "loss": 5.1794,
      "mean_token_accuracy": 0.16493744403123856,
      "num_tokens": 27388221.0,
      "step": 12500
    },
    {
      "entropy": 5.5570111751556395,
      "epoch": 1.2410678840809846,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004853927618565694,
      "loss": 5.1765,
      "mean_token_accuracy": 0.16697217226028443,
      "num_tokens": 27400497.0,
      "step": 12505
    },
    {
      "entropy": 5.483710765838623,
      "epoch": 1.241564112743152,
      "grad_norm": 1.0,
      "learning_rate": 0.00048538020092552626,
      "loss": 5.0649,
      "mean_token_accuracy": 0.17475847899913788,
      "num_tokens": 27410700.0,
      "step": 12510
    },
    {
      "entropy": 5.513055229187012,
      "epoch": 1.2420603414053195,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048536763477750747,
      "loss": 5.2065,
      "mean_token_accuracy": 0.17092445492744446,
      "num_tokens": 27421286.0,
      "step": 12515
    },
    {
      "entropy": 5.579060459136963,
      "epoch": 1.242556570067487,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004853550634128246,
      "loss": 5.2226,
      "mean_token_accuracy": 0.1669715538620949,
      "num_tokens": 27433922.0,
      "step": 12520
    },
    {
      "entropy": 5.494808101654053,
      "epoch": 1.2430527987296547,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004853424868317895,
      "loss": 5.1517,
      "mean_token_accuracy": 0.16734260618686675,
      "num_tokens": 27444684.0,
      "step": 12525
    },
    {
      "entropy": 5.490879917144776,
      "epoch": 1.2435490273918222,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004853299050347139,
      "loss": 5.1393,
      "mean_token_accuracy": 0.16786016672849655,
      "num_tokens": 27455494.0,
      "step": 12530
    },
    {
      "entropy": 5.602056550979614,
      "epoch": 1.2440452560539896,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048531731802190994,
      "loss": 5.1743,
      "mean_token_accuracy": 0.167215932905674,
      "num_tokens": 27464927.0,
      "step": 12535
    },
    {
      "entropy": 5.576707363128662,
      "epoch": 1.244541484716157,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004853047257936897,
      "loss": 5.182,
      "mean_token_accuracy": 0.16496121436357497,
      "num_tokens": 27476228.0,
      "step": 12540
    },
    {
      "entropy": 5.550760507583618,
      "epoch": 1.2450377133783248,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004852921283503653,
      "loss": 5.1847,
      "mean_token_accuracy": 0.17001437842845918,
      "num_tokens": 27488835.0,
      "step": 12545
    },
    {
      "entropy": 5.517337465286255,
      "epoch": 1.2455339420404923,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004852795256922494,
      "loss": 5.1586,
      "mean_token_accuracy": 0.16673565804958343,
      "num_tokens": 27501481.0,
      "step": 12550
    },
    {
      "entropy": 5.536225080490112,
      "epoch": 1.2460301707026598,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048526691781965425,
      "loss": 5.1774,
      "mean_token_accuracy": 0.16835180521011353,
      "num_tokens": 27512234.0,
      "step": 12555
    },
    {
      "entropy": 5.605270719528198,
      "epoch": 1.2465263993648272,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048525430473289274,
      "loss": 5.2305,
      "mean_token_accuracy": 0.16284125298261642,
      "num_tokens": 27523756.0,
      "step": 12560
    },
    {
      "entropy": 5.5720970153808596,
      "epoch": 1.247022628026995,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048524168643227747,
      "loss": 5.2201,
      "mean_token_accuracy": 0.1646725729107857,
      "num_tokens": 27534579.0,
      "step": 12565
    },
    {
      "entropy": 5.514043283462525,
      "epoch": 1.2475188566891624,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048522906291812145,
      "loss": 5.2538,
      "mean_token_accuracy": 0.16246421188116072,
      "num_tokens": 27546420.0,
      "step": 12570
    },
    {
      "entropy": 5.549548816680908,
      "epoch": 1.2480150853513299,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048521643419073763,
      "loss": 5.119,
      "mean_token_accuracy": 0.16686424612998962,
      "num_tokens": 27557701.0,
      "step": 12575
    },
    {
      "entropy": 5.480019235610962,
      "epoch": 1.2485113140134974,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004852038002504392,
      "loss": 5.1447,
      "mean_token_accuracy": 0.17416032105684282,
      "num_tokens": 27568592.0,
      "step": 12580
    },
    {
      "entropy": 5.536455869674683,
      "epoch": 1.249007542675665,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048519116109753955,
      "loss": 5.177,
      "mean_token_accuracy": 0.1732865184545517,
      "num_tokens": 27579286.0,
      "step": 12585
    },
    {
      "entropy": 5.517602252960205,
      "epoch": 1.2495037713378325,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048517851673235196,
      "loss": 5.1178,
      "mean_token_accuracy": 0.16958713978528978,
      "num_tokens": 27590031.0,
      "step": 12590
    },
    {
      "entropy": 5.517959785461426,
      "epoch": 1.25,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004851658671551901,
      "loss": 5.1522,
      "mean_token_accuracy": 0.16750041544437408,
      "num_tokens": 27600999.0,
      "step": 12595
    },
    {
      "entropy": 5.492237997055054,
      "epoch": 1.2504962286621675,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048515321236636754,
      "loss": 5.1472,
      "mean_token_accuracy": 0.16664251834154128,
      "num_tokens": 27613394.0,
      "step": 12600
    },
    {
      "entropy": 5.505588483810425,
      "epoch": 1.250992457324335,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004851405523661982,
      "loss": 5.1262,
      "mean_token_accuracy": 0.17925361543893814,
      "num_tokens": 27624819.0,
      "step": 12605
    },
    {
      "entropy": 5.46622142791748,
      "epoch": 1.2514886859865026,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048512788715499597,
      "loss": 5.1584,
      "mean_token_accuracy": 0.17537716329097747,
      "num_tokens": 27635993.0,
      "step": 12610
    },
    {
      "entropy": 5.516826629638672,
      "epoch": 1.2519849146486701,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004851152167330749,
      "loss": 5.2176,
      "mean_token_accuracy": 0.16609541624784468,
      "num_tokens": 27647298.0,
      "step": 12615
    },
    {
      "entropy": 5.586860036849975,
      "epoch": 1.2524811433108376,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048510254110074923,
      "loss": 5.1887,
      "mean_token_accuracy": 0.17025675028562545,
      "num_tokens": 27659381.0,
      "step": 12620
    },
    {
      "entropy": 5.547328615188599,
      "epoch": 1.2529773719730053,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004850898602583333,
      "loss": 5.175,
      "mean_token_accuracy": 0.16522342711687088,
      "num_tokens": 27671122.0,
      "step": 12625
    },
    {
      "entropy": 5.532905769348145,
      "epoch": 1.2534736006351728,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048507717420614153,
      "loss": 5.1663,
      "mean_token_accuracy": 0.16918721348047255,
      "num_tokens": 27683258.0,
      "step": 12630
    },
    {
      "entropy": 5.427686595916748,
      "epoch": 1.2539698292973402,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004850644829444884,
      "loss": 5.031,
      "mean_token_accuracy": 0.1729172334074974,
      "num_tokens": 27693116.0,
      "step": 12635
    },
    {
      "entropy": 5.469030237197876,
      "epoch": 1.2544660579595077,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048505178647368887,
      "loss": 5.1263,
      "mean_token_accuracy": 0.1712686911225319,
      "num_tokens": 27704859.0,
      "step": 12640
    },
    {
      "entropy": 5.478932428359985,
      "epoch": 1.2549622866216752,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004850390847940576,
      "loss": 5.0782,
      "mean_token_accuracy": 0.1733901709318161,
      "num_tokens": 27715854.0,
      "step": 12645
    },
    {
      "entropy": 5.507164812088012,
      "epoch": 1.2554585152838427,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004850263779059097,
      "loss": 5.1127,
      "mean_token_accuracy": 0.17149351239204408,
      "num_tokens": 27726398.0,
      "step": 12650
    },
    {
      "entropy": 5.538310480117798,
      "epoch": 1.2559547439460104,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004850136658095601,
      "loss": 5.1627,
      "mean_token_accuracy": 0.17338285744190216,
      "num_tokens": 27737337.0,
      "step": 12655
    },
    {
      "entropy": 5.516957759857178,
      "epoch": 1.2564509726081778,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048500094850532417,
      "loss": 5.1781,
      "mean_token_accuracy": 0.17100569158792495,
      "num_tokens": 27748593.0,
      "step": 12660
    },
    {
      "entropy": 5.572302961349488,
      "epoch": 1.2569472012703453,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004849882259935172,
      "loss": 5.1399,
      "mean_token_accuracy": 0.1682426154613495,
      "num_tokens": 27760908.0,
      "step": 12665
    },
    {
      "entropy": 5.55339789390564,
      "epoch": 1.257443429932513,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004849754982744548,
      "loss": 5.2207,
      "mean_token_accuracy": 0.16648263186216355,
      "num_tokens": 27772104.0,
      "step": 12670
    },
    {
      "entropy": 5.467471742630005,
      "epoch": 1.2579396585946805,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004849627653484524,
      "loss": 5.1654,
      "mean_token_accuracy": 0.17432896196842193,
      "num_tokens": 27781654.0,
      "step": 12675
    },
    {
      "entropy": 5.527649021148681,
      "epoch": 1.258435887256848,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004849500272158259,
      "loss": 5.1492,
      "mean_token_accuracy": 0.17107137888669968,
      "num_tokens": 27793026.0,
      "step": 12680
    },
    {
      "entropy": 5.543030500411987,
      "epoch": 1.2589321159190154,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004849372838768911,
      "loss": 5.1253,
      "mean_token_accuracy": 0.17380276918411255,
      "num_tokens": 27803447.0,
      "step": 12685
    },
    {
      "entropy": 5.6290388107299805,
      "epoch": 1.259428344581183,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000484924535331964,
      "loss": 5.1359,
      "mean_token_accuracy": 0.1681838095188141,
      "num_tokens": 27814632.0,
      "step": 12690
    },
    {
      "entropy": 5.545263957977295,
      "epoch": 1.2599245732433506,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048491178158136086,
      "loss": 5.2237,
      "mean_token_accuracy": 0.1642076313495636,
      "num_tokens": 27826582.0,
      "step": 12695
    },
    {
      "entropy": 5.516428422927857,
      "epoch": 1.260420801905518,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004848990226253978,
      "loss": 5.0708,
      "mean_token_accuracy": 0.1688187152147293,
      "num_tokens": 27838065.0,
      "step": 12700
    },
    {
      "entropy": 5.484880828857422,
      "epoch": 1.2609170305676856,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048488625846439135,
      "loss": 5.1378,
      "mean_token_accuracy": 0.17219123542308806,
      "num_tokens": 27848182.0,
      "step": 12705
    },
    {
      "entropy": 5.576004219055176,
      "epoch": 1.261413259229853,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004848734890986578,
      "loss": 5.1405,
      "mean_token_accuracy": 0.1705386146903038,
      "num_tokens": 27858524.0,
      "step": 12710
    },
    {
      "entropy": 5.535070991516113,
      "epoch": 1.2619094878920207,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004848607145285141,
      "loss": 5.0627,
      "mean_token_accuracy": 0.17471428215503693,
      "num_tokens": 27868588.0,
      "step": 12715
    },
    {
      "entropy": 5.494311952590943,
      "epoch": 1.2624057165541882,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004848479347542768,
      "loss": 5.0815,
      "mean_token_accuracy": 0.1724834606051445,
      "num_tokens": 27878829.0,
      "step": 12720
    },
    {
      "entropy": 5.500310945510864,
      "epoch": 1.2629019452163557,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000484835149776263,
      "loss": 5.1559,
      "mean_token_accuracy": 0.17567149102687835,
      "num_tokens": 27889872.0,
      "step": 12725
    },
    {
      "entropy": 5.510600996017456,
      "epoch": 1.2633981738785232,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048482235959478953,
      "loss": 5.2206,
      "mean_token_accuracy": 0.16926396936178206,
      "num_tokens": 27900567.0,
      "step": 12730
    },
    {
      "entropy": 5.555635976791382,
      "epoch": 1.2638944025406906,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004848095642101737,
      "loss": 5.1013,
      "mean_token_accuracy": 0.17302893251180648,
      "num_tokens": 27911594.0,
      "step": 12735
    },
    {
      "entropy": 5.600049161911011,
      "epoch": 1.2643906312028583,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048479676362273277,
      "loss": 5.1975,
      "mean_token_accuracy": 0.16255032867193223,
      "num_tokens": 27923694.0,
      "step": 12740
    },
    {
      "entropy": 5.554832839965821,
      "epoch": 1.2648868598650258,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048478395783278425,
      "loss": 5.2844,
      "mean_token_accuracy": 0.15380905121564864,
      "num_tokens": 27934168.0,
      "step": 12745
    },
    {
      "entropy": 5.585643911361695,
      "epoch": 1.2653830885271933,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004847711468406455,
      "loss": 5.171,
      "mean_token_accuracy": 0.16681780964136123,
      "num_tokens": 27944401.0,
      "step": 12750
    },
    {
      "entropy": 5.479246711730957,
      "epoch": 1.265879317189361,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048475833064663443,
      "loss": 5.0627,
      "mean_token_accuracy": 0.17939431369304656,
      "num_tokens": 27956219.0,
      "step": 12755
    },
    {
      "entropy": 5.5214111328125,
      "epoch": 1.2663755458515285,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004847455092510687,
      "loss": 5.1664,
      "mean_token_accuracy": 0.17302629798650743,
      "num_tokens": 27967457.0,
      "step": 12760
    },
    {
      "entropy": 5.510754060745239,
      "epoch": 1.266871774513696,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004847326826542663,
      "loss": 5.1747,
      "mean_token_accuracy": 0.16621131598949432,
      "num_tokens": 27978287.0,
      "step": 12765
    },
    {
      "entropy": 5.55013256072998,
      "epoch": 1.2673680031758634,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048471985085654537,
      "loss": 5.1987,
      "mean_token_accuracy": 0.17439829111099242,
      "num_tokens": 27989703.0,
      "step": 12770
    },
    {
      "entropy": 5.536842489242554,
      "epoch": 1.2678642318380309,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048470701385822394,
      "loss": 5.1628,
      "mean_token_accuracy": 0.1690773770213127,
      "num_tokens": 28002071.0,
      "step": 12775
    },
    {
      "entropy": 5.555324363708496,
      "epoch": 1.2683604605001986,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048469417165962047,
      "loss": 5.2169,
      "mean_token_accuracy": 0.16876873821020127,
      "num_tokens": 28014517.0,
      "step": 12780
    },
    {
      "entropy": 5.541184186935425,
      "epoch": 1.268856689162366,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004846813242610535,
      "loss": 5.1695,
      "mean_token_accuracy": 0.17397904694080352,
      "num_tokens": 28025135.0,
      "step": 12785
    },
    {
      "entropy": 5.485084056854248,
      "epoch": 1.2693529178245335,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004846684716628414,
      "loss": 5.1031,
      "mean_token_accuracy": 0.1720394656062126,
      "num_tokens": 28036105.0,
      "step": 12790
    },
    {
      "entropy": 5.490474700927734,
      "epoch": 1.269849146486701,
      "grad_norm": 0.96875,
      "learning_rate": 0.000484655613865303,
      "loss": 5.1457,
      "mean_token_accuracy": 0.17245976030826568,
      "num_tokens": 28048825.0,
      "step": 12795
    },
    {
      "entropy": 5.5143452167510985,
      "epoch": 1.2703453751488687,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048464275086875717,
      "loss": 5.0982,
      "mean_token_accuracy": 0.17062480598688126,
      "num_tokens": 28060408.0,
      "step": 12800
    },
    {
      "entropy": 5.554420852661133,
      "epoch": 1.2708416038110362,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004846298826735229,
      "loss": 5.18,
      "mean_token_accuracy": 0.1619243785738945,
      "num_tokens": 28072179.0,
      "step": 12805
    },
    {
      "entropy": 5.509398078918457,
      "epoch": 1.2713378324732036,
      "grad_norm": 1.125,
      "learning_rate": 0.0004846170092799192,
      "loss": 5.1451,
      "mean_token_accuracy": 0.16915974915027618,
      "num_tokens": 28083825.0,
      "step": 12810
    },
    {
      "entropy": 5.6818922996521,
      "epoch": 1.2718340611353711,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048460413068826536,
      "loss": 5.3056,
      "mean_token_accuracy": 0.16124261915683746,
      "num_tokens": 28095136.0,
      "step": 12815
    },
    {
      "entropy": 5.570952320098877,
      "epoch": 1.2723302897975386,
      "grad_norm": 1.0,
      "learning_rate": 0.0004845912468988807,
      "loss": 5.1261,
      "mean_token_accuracy": 0.16855156868696214,
      "num_tokens": 28104763.0,
      "step": 12820
    },
    {
      "entropy": 5.528598594665527,
      "epoch": 1.2728265184597063,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004845783579120849,
      "loss": 5.1798,
      "mean_token_accuracy": 0.16893855780363082,
      "num_tokens": 28115771.0,
      "step": 12825
    },
    {
      "entropy": 5.492273235321045,
      "epoch": 1.2733227471218738,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004845654637281973,
      "loss": 5.1078,
      "mean_token_accuracy": 0.16819423139095308,
      "num_tokens": 28128039.0,
      "step": 12830
    },
    {
      "entropy": 5.486709690093994,
      "epoch": 1.2738189757840412,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048455256434753774,
      "loss": 5.1206,
      "mean_token_accuracy": 0.1762232467532158,
      "num_tokens": 28139163.0,
      "step": 12835
    },
    {
      "entropy": 5.484812641143799,
      "epoch": 1.274315204446209,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004845396597704262,
      "loss": 5.1585,
      "mean_token_accuracy": 0.16834336072206496,
      "num_tokens": 28150625.0,
      "step": 12840
    },
    {
      "entropy": 5.569616365432739,
      "epoch": 1.2748114331083764,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048452674999718257,
      "loss": 5.2309,
      "mean_token_accuracy": 0.16711384654045106,
      "num_tokens": 28163072.0,
      "step": 12845
    },
    {
      "entropy": 5.623217678070068,
      "epoch": 1.2753076617705439,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048451383502812705,
      "loss": 5.2518,
      "mean_token_accuracy": 0.16317146867513657,
      "num_tokens": 28174192.0,
      "step": 12850
    },
    {
      "entropy": 5.48516263961792,
      "epoch": 1.2758038904327114,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048450091486357987,
      "loss": 5.1015,
      "mean_token_accuracy": 0.17857202589511872,
      "num_tokens": 28184477.0,
      "step": 12855
    },
    {
      "entropy": 5.472021150588989,
      "epoch": 1.2763001190948788,
      "grad_norm": 0.875,
      "learning_rate": 0.00048448798950386147,
      "loss": 5.0274,
      "mean_token_accuracy": 0.17695969194173813,
      "num_tokens": 28194552.0,
      "step": 12860
    },
    {
      "entropy": 5.510885238647461,
      "epoch": 1.2767963477570463,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048447505894929233,
      "loss": 5.1973,
      "mean_token_accuracy": 0.1698785126209259,
      "num_tokens": 28205712.0,
      "step": 12865
    },
    {
      "entropy": 5.549971055984497,
      "epoch": 1.277292576419214,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048446212320019306,
      "loss": 5.1305,
      "mean_token_accuracy": 0.17509013861417771,
      "num_tokens": 28217246.0,
      "step": 12870
    },
    {
      "entropy": 5.52444543838501,
      "epoch": 1.2777888050813815,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004844491822568846,
      "loss": 5.1486,
      "mean_token_accuracy": 0.16876955032348634,
      "num_tokens": 28227933.0,
      "step": 12875
    },
    {
      "entropy": 5.517069387435913,
      "epoch": 1.278285033743549,
      "grad_norm": 0.84375,
      "learning_rate": 0.00048443623611968764,
      "loss": 5.1709,
      "mean_token_accuracy": 0.17192982733249665,
      "num_tokens": 28239692.0,
      "step": 12880
    },
    {
      "entropy": 5.503536605834961,
      "epoch": 1.2787812624057167,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048442328478892334,
      "loss": 5.1422,
      "mean_token_accuracy": 0.1741974472999573,
      "num_tokens": 28249815.0,
      "step": 12885
    },
    {
      "entropy": 5.57446665763855,
      "epoch": 1.2792774910678841,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004844103282649128,
      "loss": 5.171,
      "mean_token_accuracy": 0.16653064787387847,
      "num_tokens": 28260385.0,
      "step": 12890
    },
    {
      "entropy": 5.502856588363647,
      "epoch": 1.2797737197300516,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004843973665479774,
      "loss": 5.1807,
      "mean_token_accuracy": 0.16114374548196791,
      "num_tokens": 28271858.0,
      "step": 12895
    },
    {
      "entropy": 5.567032337188721,
      "epoch": 1.280269948392219,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048438439963843853,
      "loss": 5.1788,
      "mean_token_accuracy": 0.1681448847055435,
      "num_tokens": 28283988.0,
      "step": 12900
    },
    {
      "entropy": 5.5415338516235355,
      "epoch": 1.2807661770543866,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048437142753661775,
      "loss": 5.1976,
      "mean_token_accuracy": 0.17178643643856048,
      "num_tokens": 28293281.0,
      "step": 12905
    },
    {
      "entropy": 5.614867639541626,
      "epoch": 1.2812624057165543,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004843584502428367,
      "loss": 5.1541,
      "mean_token_accuracy": 0.16519233137369155,
      "num_tokens": 28304468.0,
      "step": 12910
    },
    {
      "entropy": 5.446540594100952,
      "epoch": 1.2817586343787217,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004843454677574173,
      "loss": 5.1069,
      "mean_token_accuracy": 0.1761067196726799,
      "num_tokens": 28315864.0,
      "step": 12915
    },
    {
      "entropy": 5.501544523239136,
      "epoch": 1.2822548630408892,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048433248008068123,
      "loss": 5.1761,
      "mean_token_accuracy": 0.1747245654463768,
      "num_tokens": 28327212.0,
      "step": 12920
    },
    {
      "entropy": 5.572963809967041,
      "epoch": 1.282751091703057,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048431948721295085,
      "loss": 5.1713,
      "mean_token_accuracy": 0.17358099520206452,
      "num_tokens": 28339256.0,
      "step": 12925
    },
    {
      "entropy": 5.55533127784729,
      "epoch": 1.2832473203652244,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004843064891545482,
      "loss": 5.1795,
      "mean_token_accuracy": 0.17052376717329026,
      "num_tokens": 28350196.0,
      "step": 12930
    },
    {
      "entropy": 5.530222606658936,
      "epoch": 1.2837435490273919,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004842934859057956,
      "loss": 5.1561,
      "mean_token_accuracy": 0.16836240887641907,
      "num_tokens": 28361241.0,
      "step": 12935
    },
    {
      "entropy": 5.589635801315308,
      "epoch": 1.2842397776895593,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004842804774670156,
      "loss": 5.2759,
      "mean_token_accuracy": 0.16310614496469497,
      "num_tokens": 28373369.0,
      "step": 12940
    },
    {
      "entropy": 5.547092342376709,
      "epoch": 1.2847360063517268,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048426746383853066,
      "loss": 5.1043,
      "mean_token_accuracy": 0.17407662719488143,
      "num_tokens": 28384970.0,
      "step": 12945
    },
    {
      "entropy": 5.540127944946289,
      "epoch": 1.2852322350138943,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048425444502066354,
      "loss": 5.1157,
      "mean_token_accuracy": 0.17156493216753005,
      "num_tokens": 28396469.0,
      "step": 12950
    },
    {
      "entropy": 5.512262153625488,
      "epoch": 1.285728463676062,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048424142101373706,
      "loss": 5.1805,
      "mean_token_accuracy": 0.169954714179039,
      "num_tokens": 28406975.0,
      "step": 12955
    },
    {
      "entropy": 5.466426658630371,
      "epoch": 1.2862246923382294,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048422839181807426,
      "loss": 5.0496,
      "mean_token_accuracy": 0.18006895929574968,
      "num_tokens": 28418285.0,
      "step": 12960
    },
    {
      "entropy": 5.495448303222656,
      "epoch": 1.286720921000397,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004842153574339982,
      "loss": 5.1984,
      "mean_token_accuracy": 0.16441502571105956,
      "num_tokens": 28429769.0,
      "step": 12965
    },
    {
      "entropy": 5.499081182479858,
      "epoch": 1.2872171496625646,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048420231786183216,
      "loss": 5.1408,
      "mean_token_accuracy": 0.17683082073926926,
      "num_tokens": 28441188.0,
      "step": 12970
    },
    {
      "entropy": 5.597442388534546,
      "epoch": 1.287713378324732,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004841892731018994,
      "loss": 5.2034,
      "mean_token_accuracy": 0.16595323234796525,
      "num_tokens": 28453504.0,
      "step": 12975
    },
    {
      "entropy": 5.568818998336792,
      "epoch": 1.2882096069868996,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004841762231545233,
      "loss": 5.186,
      "mean_token_accuracy": 0.1743595689535141,
      "num_tokens": 28465930.0,
      "step": 12980
    },
    {
      "entropy": 5.509972286224365,
      "epoch": 1.288705835649067,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048416316802002776,
      "loss": 5.1023,
      "mean_token_accuracy": 0.17614503800868989,
      "num_tokens": 28477803.0,
      "step": 12985
    },
    {
      "entropy": 5.570358085632324,
      "epoch": 1.2892020643112345,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048415010769873625,
      "loss": 5.2103,
      "mean_token_accuracy": 0.1656086727976799,
      "num_tokens": 28488556.0,
      "step": 12990
    },
    {
      "entropy": 5.561723375320435,
      "epoch": 1.2896982929734022,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004841370421909728,
      "loss": 5.1122,
      "mean_token_accuracy": 0.17091791033744813,
      "num_tokens": 28499275.0,
      "step": 12995
    },
    {
      "entropy": 5.56761794090271,
      "epoch": 1.2901945216355697,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048412397149706133,
      "loss": 5.149,
      "mean_token_accuracy": 0.17340620160102843,
      "num_tokens": 28511327.0,
      "step": 13000
    },
    {
      "entropy": 5.521489238739013,
      "epoch": 1.2906907502977372,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000484110895617326,
      "loss": 5.1802,
      "mean_token_accuracy": 0.17062077671289444,
      "num_tokens": 28523937.0,
      "step": 13005
    },
    {
      "entropy": 5.437697410583496,
      "epoch": 1.2911869789599046,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048409781455209104,
      "loss": 5.0882,
      "mean_token_accuracy": 0.17024388015270234,
      "num_tokens": 28534028.0,
      "step": 13010
    },
    {
      "entropy": 5.588116025924682,
      "epoch": 1.2916832076220723,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004840847283016809,
      "loss": 5.1797,
      "mean_token_accuracy": 0.15891446620225907,
      "num_tokens": 28545480.0,
      "step": 13015
    },
    {
      "entropy": 5.5265740871429445,
      "epoch": 1.2921794362842398,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048407163686642,
      "loss": 5.1218,
      "mean_token_accuracy": 0.17879820466041565,
      "num_tokens": 28555401.0,
      "step": 13020
    },
    {
      "entropy": 5.5234174728393555,
      "epoch": 1.2926756649464073,
      "grad_norm": 1.078125,
      "learning_rate": 0.000484058540246633,
      "loss": 5.1953,
      "mean_token_accuracy": 0.17124780863523484,
      "num_tokens": 28565497.0,
      "step": 13025
    },
    {
      "entropy": 5.5675750255584715,
      "epoch": 1.2931718936085748,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004840454384426447,
      "loss": 5.2786,
      "mean_token_accuracy": 0.1606290251016617,
      "num_tokens": 28576186.0,
      "step": 13030
    },
    {
      "entropy": 5.598295259475708,
      "epoch": 1.2936681222707422,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004840323314547799,
      "loss": 5.1463,
      "mean_token_accuracy": 0.1722848042845726,
      "num_tokens": 28586202.0,
      "step": 13035
    },
    {
      "entropy": 5.519181776046753,
      "epoch": 1.29416435093291,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048401921928336385,
      "loss": 5.1141,
      "mean_token_accuracy": 0.16592354327440262,
      "num_tokens": 28596486.0,
      "step": 13040
    },
    {
      "entropy": 5.55699315071106,
      "epoch": 1.2946605795950774,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004840061019287215,
      "loss": 5.1574,
      "mean_token_accuracy": 0.171697960793972,
      "num_tokens": 28607934.0,
      "step": 13045
    },
    {
      "entropy": 5.614859914779663,
      "epoch": 1.2951568082572449,
      "grad_norm": 1.09375,
      "learning_rate": 0.00048399297939117817,
      "loss": 5.1934,
      "mean_token_accuracy": 0.16277693808078766,
      "num_tokens": 28617673.0,
      "step": 13050
    },
    {
      "entropy": 5.6102077007293705,
      "epoch": 1.2956530369194126,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004839798516710593,
      "loss": 5.1832,
      "mean_token_accuracy": 0.17069528847932816,
      "num_tokens": 28629811.0,
      "step": 13055
    },
    {
      "entropy": 5.596235704421997,
      "epoch": 1.29614926558158,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004839667187686904,
      "loss": 5.2612,
      "mean_token_accuracy": 0.1647556334733963,
      "num_tokens": 28640371.0,
      "step": 13060
    },
    {
      "entropy": 5.548670816421509,
      "epoch": 1.2966454942437475,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004839535806843972,
      "loss": 5.1893,
      "mean_token_accuracy": 0.16967585012316705,
      "num_tokens": 28651904.0,
      "step": 13065
    },
    {
      "entropy": 5.488644027709961,
      "epoch": 1.297141722905915,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004839404374185054,
      "loss": 5.1144,
      "mean_token_accuracy": 0.1673978701233864,
      "num_tokens": 28662241.0,
      "step": 13070
    },
    {
      "entropy": 5.491305589675903,
      "epoch": 1.2976379515680825,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000483927288971341,
      "loss": 5.0856,
      "mean_token_accuracy": 0.17415677905082702,
      "num_tokens": 28672791.0,
      "step": 13075
    },
    {
      "entropy": 5.525943326950073,
      "epoch": 1.2981341802302502,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004839141353432301,
      "loss": 5.2222,
      "mean_token_accuracy": 0.16960235834121704,
      "num_tokens": 28684218.0,
      "step": 13080
    },
    {
      "entropy": 5.547878217697144,
      "epoch": 1.2986304088924177,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004839009765344987,
      "loss": 5.153,
      "mean_token_accuracy": 0.1657433107495308,
      "num_tokens": 28694690.0,
      "step": 13085
    },
    {
      "entropy": 5.523418760299682,
      "epoch": 1.2991266375545851,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048388781254547324,
      "loss": 5.1449,
      "mean_token_accuracy": 0.17216846644878386,
      "num_tokens": 28705890.0,
      "step": 13090
    },
    {
      "entropy": 5.548996448516846,
      "epoch": 1.2996228662167526,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048387464337648026,
      "loss": 5.1767,
      "mean_token_accuracy": 0.17067211419343947,
      "num_tokens": 28716935.0,
      "step": 13095
    },
    {
      "entropy": 5.563096284866333,
      "epoch": 1.3001190948789203,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048386146902784606,
      "loss": 5.1636,
      "mean_token_accuracy": 0.17251261323690414,
      "num_tokens": 28728053.0,
      "step": 13100
    },
    {
      "entropy": 5.48268313407898,
      "epoch": 1.3006153235410878,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048384828949989757,
      "loss": 5.1171,
      "mean_token_accuracy": 0.17329630851745606,
      "num_tokens": 28738940.0,
      "step": 13105
    },
    {
      "entropy": 5.572353172302246,
      "epoch": 1.3011115522032553,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048383510479296153,
      "loss": 5.1976,
      "mean_token_accuracy": 0.17301127016544343,
      "num_tokens": 28750259.0,
      "step": 13110
    },
    {
      "entropy": 5.55026683807373,
      "epoch": 1.3016077808654227,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004838219149073649,
      "loss": 5.1369,
      "mean_token_accuracy": 0.17226935178041458,
      "num_tokens": 28760119.0,
      "step": 13115
    },
    {
      "entropy": 5.564360618591309,
      "epoch": 1.3021040095275902,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004838087198434347,
      "loss": 5.1804,
      "mean_token_accuracy": 0.16659301668405532,
      "num_tokens": 28772022.0,
      "step": 13120
    },
    {
      "entropy": 5.519193410873413,
      "epoch": 1.302600238189758,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004837955196014982,
      "loss": 5.1281,
      "mean_token_accuracy": 0.16583479195833206,
      "num_tokens": 28782935.0,
      "step": 13125
    },
    {
      "entropy": 5.583038854598999,
      "epoch": 1.3030964668519254,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048378231418188275,
      "loss": 5.2522,
      "mean_token_accuracy": 0.17253718376159669,
      "num_tokens": 28793604.0,
      "step": 13130
    },
    {
      "entropy": 5.563717365264893,
      "epoch": 1.3035926955140928,
      "grad_norm": 1.0,
      "learning_rate": 0.00048376910358491576,
      "loss": 5.2046,
      "mean_token_accuracy": 0.16530329287052153,
      "num_tokens": 28803494.0,
      "step": 13135
    },
    {
      "entropy": 5.524064350128174,
      "epoch": 1.3040889241762605,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004837558878109249,
      "loss": 5.1686,
      "mean_token_accuracy": 0.16294043362140656,
      "num_tokens": 28816232.0,
      "step": 13140
    },
    {
      "entropy": 5.543507099151611,
      "epoch": 1.304585152838428,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004837426668602378,
      "loss": 5.1229,
      "mean_token_accuracy": 0.17397501468658447,
      "num_tokens": 28827056.0,
      "step": 13145
    },
    {
      "entropy": 5.488945960998535,
      "epoch": 1.3050813815005955,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004837294407331825,
      "loss": 5.1199,
      "mean_token_accuracy": 0.17302099019289016,
      "num_tokens": 28837217.0,
      "step": 13150
    },
    {
      "entropy": 5.457980966567993,
      "epoch": 1.305577610162763,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004837162094300867,
      "loss": 5.0278,
      "mean_token_accuracy": 0.17647732198238372,
      "num_tokens": 28846838.0,
      "step": 13155
    },
    {
      "entropy": 5.524992418289185,
      "epoch": 1.3060738388249304,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004837029729512787,
      "loss": 5.076,
      "mean_token_accuracy": 0.1775840237736702,
      "num_tokens": 28856269.0,
      "step": 13160
    },
    {
      "entropy": 5.53138484954834,
      "epoch": 1.306570067487098,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048368973129708667,
      "loss": 5.1132,
      "mean_token_accuracy": 0.17624222338199616,
      "num_tokens": 28867142.0,
      "step": 13165
    },
    {
      "entropy": 5.518300867080688,
      "epoch": 1.3070662961492656,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00048367648446783894,
      "loss": 5.0767,
      "mean_token_accuracy": 0.17221258133649825,
      "num_tokens": 28877346.0,
      "step": 13170
    },
    {
      "entropy": 5.5300086498260494,
      "epoch": 1.307562524811433,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004836632324638642,
      "loss": 5.0827,
      "mean_token_accuracy": 0.1707124665379524,
      "num_tokens": 28888425.0,
      "step": 13175
    },
    {
      "entropy": 5.5115738868713375,
      "epoch": 1.3080587534736006,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048364997528549077,
      "loss": 5.1278,
      "mean_token_accuracy": 0.16951045542955398,
      "num_tokens": 28899575.0,
      "step": 13180
    },
    {
      "entropy": 5.497907924652099,
      "epoch": 1.3085549821357683,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048363671293304763,
      "loss": 5.1094,
      "mean_token_accuracy": 0.1746928110718727,
      "num_tokens": 28910769.0,
      "step": 13185
    },
    {
      "entropy": 5.532970237731933,
      "epoch": 1.3090512107979357,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004836234454068636,
      "loss": 5.1157,
      "mean_token_accuracy": 0.1782885417342186,
      "num_tokens": 28921224.0,
      "step": 13190
    },
    {
      "entropy": 5.497860527038574,
      "epoch": 1.3095474394601032,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048361017270726764,
      "loss": 5.1157,
      "mean_token_accuracy": 0.17214779704809188,
      "num_tokens": 28932585.0,
      "step": 13195
    },
    {
      "entropy": 5.498167324066162,
      "epoch": 1.3100436681222707,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004835968948345889,
      "loss": 5.1642,
      "mean_token_accuracy": 0.1656687542796135,
      "num_tokens": 28943471.0,
      "step": 13200
    },
    {
      "entropy": 5.475327730178833,
      "epoch": 1.3105398967844382,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048358361178915666,
      "loss": 5.0728,
      "mean_token_accuracy": 0.18596817404031754,
      "num_tokens": 28956336.0,
      "step": 13205
    },
    {
      "entropy": 5.603514766693115,
      "epoch": 1.3110361254466059,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004835703235713003,
      "loss": 5.2125,
      "mean_token_accuracy": 0.1666790872812271,
      "num_tokens": 28966689.0,
      "step": 13210
    },
    {
      "entropy": 5.589122915267945,
      "epoch": 1.3115323541087733,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048355703018134935,
      "loss": 5.3069,
      "mean_token_accuracy": 0.1742893025279045,
      "num_tokens": 28978548.0,
      "step": 13215
    },
    {
      "entropy": 5.587890911102295,
      "epoch": 1.3120285827709408,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004835437316196335,
      "loss": 5.2115,
      "mean_token_accuracy": 0.16657934039831163,
      "num_tokens": 28988899.0,
      "step": 13220
    },
    {
      "entropy": 5.508882236480713,
      "epoch": 1.3125248114331085,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004835304278864824,
      "loss": 5.134,
      "mean_token_accuracy": 0.17632086873054503,
      "num_tokens": 28999666.0,
      "step": 13225
    },
    {
      "entropy": 5.546355867385865,
      "epoch": 1.313021040095276,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004835171189822261,
      "loss": 5.1365,
      "mean_token_accuracy": 0.16771157681941987,
      "num_tokens": 29010185.0,
      "step": 13230
    },
    {
      "entropy": 5.495452642440796,
      "epoch": 1.3135172687574435,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004835038049071945,
      "loss": 5.1506,
      "mean_token_accuracy": 0.16959853321313859,
      "num_tokens": 29020761.0,
      "step": 13235
    },
    {
      "entropy": 5.624324655532837,
      "epoch": 1.314013497419611,
      "grad_norm": 0.875,
      "learning_rate": 0.0004834904856617178,
      "loss": 5.2992,
      "mean_token_accuracy": 0.16602130234241486,
      "num_tokens": 29031844.0,
      "step": 13240
    },
    {
      "entropy": 5.645944356918335,
      "epoch": 1.3145097260817784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048347716124612643,
      "loss": 5.2515,
      "mean_token_accuracy": 0.1639154449105263,
      "num_tokens": 29043101.0,
      "step": 13245
    },
    {
      "entropy": 5.585401725769043,
      "epoch": 1.3150059547439459,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004834638316607506,
      "loss": 5.1857,
      "mean_token_accuracy": 0.17143767178058625,
      "num_tokens": 29054016.0,
      "step": 13250
    },
    {
      "entropy": 5.46118278503418,
      "epoch": 1.3155021834061136,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048345049690592105,
      "loss": 5.1558,
      "mean_token_accuracy": 0.16566675156354904,
      "num_tokens": 29065127.0,
      "step": 13255
    },
    {
      "entropy": 5.568687295913696,
      "epoch": 1.315998412068281,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048343715698196814,
      "loss": 5.128,
      "mean_token_accuracy": 0.16824142634868622,
      "num_tokens": 29075130.0,
      "step": 13260
    },
    {
      "entropy": 5.602110385894775,
      "epoch": 1.3164946407304485,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004834238118892231,
      "loss": 5.2484,
      "mean_token_accuracy": 0.16015904396772385,
      "num_tokens": 29086405.0,
      "step": 13265
    },
    {
      "entropy": 5.580699920654297,
      "epoch": 1.3169908693926162,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004834104616280165,
      "loss": 5.1951,
      "mean_token_accuracy": 0.16180069744586945,
      "num_tokens": 29096945.0,
      "step": 13270
    },
    {
      "entropy": 5.598345518112183,
      "epoch": 1.3174870980547837,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004833971061986796,
      "loss": 5.2685,
      "mean_token_accuracy": 0.17143092602491378,
      "num_tokens": 29108275.0,
      "step": 13275
    },
    {
      "entropy": 5.546763277053833,
      "epoch": 1.3179833267169512,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048338374560154344,
      "loss": 5.1045,
      "mean_token_accuracy": 0.16976259797811508,
      "num_tokens": 29118832.0,
      "step": 13280
    },
    {
      "entropy": 5.493811464309692,
      "epoch": 1.3184795553791187,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004833703798369395,
      "loss": 5.1323,
      "mean_token_accuracy": 0.17113442569971085,
      "num_tokens": 29130720.0,
      "step": 13285
    },
    {
      "entropy": 5.5654833793640135,
      "epoch": 1.3189757840412861,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004833570089051992,
      "loss": 5.1177,
      "mean_token_accuracy": 0.16917258203029634,
      "num_tokens": 29140441.0,
      "step": 13290
    },
    {
      "entropy": 5.544336223602295,
      "epoch": 1.3194720127034538,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048334363280665393,
      "loss": 5.1756,
      "mean_token_accuracy": 0.16945491433143617,
      "num_tokens": 29151544.0,
      "step": 13295
    },
    {
      "entropy": 5.593157339096069,
      "epoch": 1.3199682413656213,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004833302515416355,
      "loss": 5.2466,
      "mean_token_accuracy": 0.16966851353645324,
      "num_tokens": 29165475.0,
      "step": 13300
    },
    {
      "entropy": 5.5261537551879885,
      "epoch": 1.3204644700277888,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048331686511047584,
      "loss": 5.0912,
      "mean_token_accuracy": 0.1693265989422798,
      "num_tokens": 29177139.0,
      "step": 13305
    },
    {
      "entropy": 5.587190103530884,
      "epoch": 1.3209606986899565,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048330347351350676,
      "loss": 5.1689,
      "mean_token_accuracy": 0.17337473332881928,
      "num_tokens": 29187812.0,
      "step": 13310
    },
    {
      "entropy": 5.503566312789917,
      "epoch": 1.321456927352124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048329007675106036,
      "loss": 5.0666,
      "mean_token_accuracy": 0.17029308080673217,
      "num_tokens": 29197932.0,
      "step": 13315
    },
    {
      "entropy": 5.472479248046875,
      "epoch": 1.3219531560142914,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00048327667482346895,
      "loss": 5.0908,
      "mean_token_accuracy": 0.17620163708925246,
      "num_tokens": 29209529.0,
      "step": 13320
    },
    {
      "entropy": 5.516754531860352,
      "epoch": 1.322449384676459,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004832632677310647,
      "loss": 5.1357,
      "mean_token_accuracy": 0.17173322290182114,
      "num_tokens": 29219027.0,
      "step": 13325
    },
    {
      "entropy": 5.556044960021973,
      "epoch": 1.3229456133386264,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048324985547418026,
      "loss": 5.1696,
      "mean_token_accuracy": 0.16752573400735854,
      "num_tokens": 29229647.0,
      "step": 13330
    },
    {
      "entropy": 5.575425386428833,
      "epoch": 1.3234418420007938,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004832364380531482,
      "loss": 5.1823,
      "mean_token_accuracy": 0.17596760541200637,
      "num_tokens": 29240263.0,
      "step": 13335
    },
    {
      "entropy": 5.576332330703735,
      "epoch": 1.3239380706629615,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048322301546830103,
      "loss": 5.2026,
      "mean_token_accuracy": 0.16378069072961807,
      "num_tokens": 29251115.0,
      "step": 13340
    },
    {
      "entropy": 5.514184141159058,
      "epoch": 1.324434299325129,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004832095877199718,
      "loss": 5.1748,
      "mean_token_accuracy": 0.16429441869258882,
      "num_tokens": 29261888.0,
      "step": 13345
    },
    {
      "entropy": 5.46731743812561,
      "epoch": 1.3249305279872965,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048319615480849347,
      "loss": 5.0739,
      "mean_token_accuracy": 0.17850470691919326,
      "num_tokens": 29271948.0,
      "step": 13350
    },
    {
      "entropy": 5.542776489257813,
      "epoch": 1.3254267566494642,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048318271673419914,
      "loss": 5.1444,
      "mean_token_accuracy": 0.16894257217645645,
      "num_tokens": 29282449.0,
      "step": 13355
    },
    {
      "entropy": 5.539471769332886,
      "epoch": 1.3259229853116317,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048316927349742207,
      "loss": 5.1662,
      "mean_token_accuracy": 0.1671816036105156,
      "num_tokens": 29294280.0,
      "step": 13360
    },
    {
      "entropy": 5.571553325653076,
      "epoch": 1.3264192139737991,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004831558250984955,
      "loss": 5.1328,
      "mean_token_accuracy": 0.1717141091823578,
      "num_tokens": 29305143.0,
      "step": 13365
    },
    {
      "entropy": 5.535297060012818,
      "epoch": 1.3269154426359666,
      "grad_norm": 1.125,
      "learning_rate": 0.00048314237153775303,
      "loss": 5.0694,
      "mean_token_accuracy": 0.17472027838230134,
      "num_tokens": 29314697.0,
      "step": 13370
    },
    {
      "entropy": 5.450375318527222,
      "epoch": 1.327411671298134,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004831289128155283,
      "loss": 5.0989,
      "mean_token_accuracy": 0.16731479465961457,
      "num_tokens": 29325257.0,
      "step": 13375
    },
    {
      "entropy": 5.537502670288086,
      "epoch": 1.3279078999603018,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048311544893215496,
      "loss": 5.1876,
      "mean_token_accuracy": 0.1664431095123291,
      "num_tokens": 29335449.0,
      "step": 13380
    },
    {
      "entropy": 5.552964925765991,
      "epoch": 1.3284041286224693,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000483101979887967,
      "loss": 5.1819,
      "mean_token_accuracy": 0.16522295773029327,
      "num_tokens": 29346898.0,
      "step": 13385
    },
    {
      "entropy": 5.596976900100708,
      "epoch": 1.3289003572846367,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048308850568329825,
      "loss": 5.1364,
      "mean_token_accuracy": 0.16628507971763612,
      "num_tokens": 29356118.0,
      "step": 13390
    },
    {
      "entropy": 5.535267114639282,
      "epoch": 1.3293965859468042,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048307502631848306,
      "loss": 5.2179,
      "mean_token_accuracy": 0.1626502111554146,
      "num_tokens": 29367597.0,
      "step": 13395
    },
    {
      "entropy": 5.480478763580322,
      "epoch": 1.329892814608972,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048306154179385546,
      "loss": 5.0871,
      "mean_token_accuracy": 0.16878528892993927,
      "num_tokens": 29377829.0,
      "step": 13400
    },
    {
      "entropy": 5.504371070861817,
      "epoch": 1.3303890432711394,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004830480521097501,
      "loss": 5.0596,
      "mean_token_accuracy": 0.1710459351539612,
      "num_tokens": 29387945.0,
      "step": 13405
    },
    {
      "entropy": 5.455507373809814,
      "epoch": 1.3308852719333069,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004830345572665012,
      "loss": 5.0287,
      "mean_token_accuracy": 0.18020383268594742,
      "num_tokens": 29398269.0,
      "step": 13410
    },
    {
      "entropy": 5.593930053710937,
      "epoch": 1.3313815005954743,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048302105726444366,
      "loss": 5.2105,
      "mean_token_accuracy": 0.16092029213905334,
      "num_tokens": 29410946.0,
      "step": 13415
    },
    {
      "entropy": 5.518677616119385,
      "epoch": 1.3318777292576418,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048300755210391205,
      "loss": 5.0567,
      "mean_token_accuracy": 0.1721974492073059,
      "num_tokens": 29421770.0,
      "step": 13420
    },
    {
      "entropy": 5.52910008430481,
      "epoch": 1.3323739579198095,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004829940417852415,
      "loss": 5.2157,
      "mean_token_accuracy": 0.16812992990016937,
      "num_tokens": 29433717.0,
      "step": 13425
    },
    {
      "entropy": 5.470255279541016,
      "epoch": 1.332870186581977,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004829805263087668,
      "loss": 5.0539,
      "mean_token_accuracy": 0.17629618048667908,
      "num_tokens": 29444382.0,
      "step": 13430
    },
    {
      "entropy": 5.564712810516357,
      "epoch": 1.3333664152441445,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048296700567482324,
      "loss": 5.1767,
      "mean_token_accuracy": 0.16681267470121383,
      "num_tokens": 29454680.0,
      "step": 13435
    },
    {
      "entropy": 5.577697801589966,
      "epoch": 1.3338626439063122,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004829534798837461,
      "loss": 5.1812,
      "mean_token_accuracy": 0.17264432609081268,
      "num_tokens": 29465065.0,
      "step": 13440
    },
    {
      "entropy": 5.579254531860352,
      "epoch": 1.3343588725684796,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004829399489358707,
      "loss": 5.1219,
      "mean_token_accuracy": 0.1692115396261215,
      "num_tokens": 29475704.0,
      "step": 13445
    },
    {
      "entropy": 5.503735017776489,
      "epoch": 1.334855101230647,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004829264128315326,
      "loss": 5.1457,
      "mean_token_accuracy": 0.16468473374843598,
      "num_tokens": 29486418.0,
      "step": 13450
    },
    {
      "entropy": 5.58544545173645,
      "epoch": 1.3353513298928146,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004829128715710676,
      "loss": 5.2362,
      "mean_token_accuracy": 0.1592664211988449,
      "num_tokens": 29498939.0,
      "step": 13455
    },
    {
      "entropy": 5.514646100997925,
      "epoch": 1.335847558554982,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004828993251548114,
      "loss": 5.0616,
      "mean_token_accuracy": 0.1790044367313385,
      "num_tokens": 29508525.0,
      "step": 13460
    },
    {
      "entropy": 5.556839656829834,
      "epoch": 1.3363437872171495,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004828857735830999,
      "loss": 5.1682,
      "mean_token_accuracy": 0.17363756895065308,
      "num_tokens": 29519962.0,
      "step": 13465
    },
    {
      "entropy": 5.493147087097168,
      "epoch": 1.3368400158793172,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004828722168562691,
      "loss": 5.1238,
      "mean_token_accuracy": 0.17261976152658462,
      "num_tokens": 29529586.0,
      "step": 13470
    },
    {
      "entropy": 5.520703887939453,
      "epoch": 1.3373362445414847,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004828586549746553,
      "loss": 5.2173,
      "mean_token_accuracy": 0.16360023617744446,
      "num_tokens": 29540670.0,
      "step": 13475
    },
    {
      "entropy": 5.56369047164917,
      "epoch": 1.3378324732036522,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004828450879385947,
      "loss": 5.1736,
      "mean_token_accuracy": 0.16526678800582886,
      "num_tokens": 29551484.0,
      "step": 13480
    },
    {
      "entropy": 5.548616695404053,
      "epoch": 1.3383287018658199,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004828315157484239,
      "loss": 5.1156,
      "mean_token_accuracy": 0.17327941507101058,
      "num_tokens": 29561596.0,
      "step": 13485
    },
    {
      "entropy": 5.56624665260315,
      "epoch": 1.3388249305279873,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004828179384044793,
      "loss": 5.1339,
      "mean_token_accuracy": 0.1695005089044571,
      "num_tokens": 29572837.0,
      "step": 13490
    },
    {
      "entropy": 5.533002471923828,
      "epoch": 1.3393211591901548,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048280435590709756,
      "loss": 5.0882,
      "mean_token_accuracy": 0.17123755514621736,
      "num_tokens": 29584095.0,
      "step": 13495
    },
    {
      "entropy": 5.530926465988159,
      "epoch": 1.3398173878523223,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004827907682566157,
      "loss": 5.1395,
      "mean_token_accuracy": 0.16533828973770143,
      "num_tokens": 29594903.0,
      "step": 13500
    },
    {
      "entropy": 5.530222988128662,
      "epoch": 1.3403136165144898,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004827771754533704,
      "loss": 5.1182,
      "mean_token_accuracy": 0.17224762886762618,
      "num_tokens": 29605199.0,
      "step": 13505
    },
    {
      "entropy": 5.5968035697937015,
      "epoch": 1.3408098451766575,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048276357749769895,
      "loss": 5.2949,
      "mean_token_accuracy": 0.16060142517089843,
      "num_tokens": 29617535.0,
      "step": 13510
    },
    {
      "entropy": 5.529836797714234,
      "epoch": 1.341306073838825,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048274997438993845,
      "loss": 5.088,
      "mean_token_accuracy": 0.171018186211586,
      "num_tokens": 29628227.0,
      "step": 13515
    },
    {
      "entropy": 5.57895073890686,
      "epoch": 1.3418023025009924,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048273636613042626,
      "loss": 5.2037,
      "mean_token_accuracy": 0.1692424848675728,
      "num_tokens": 29639112.0,
      "step": 13520
    },
    {
      "entropy": 5.455804634094238,
      "epoch": 1.3422985311631601,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004827227527194998,
      "loss": 5.0881,
      "mean_token_accuracy": 0.16489313095808028,
      "num_tokens": 29650768.0,
      "step": 13525
    },
    {
      "entropy": 5.565947532653809,
      "epoch": 1.3427947598253276,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004827091341574967,
      "loss": 5.2026,
      "mean_token_accuracy": 0.17328959852457046,
      "num_tokens": 29662388.0,
      "step": 13530
    },
    {
      "entropy": 5.551366996765137,
      "epoch": 1.343290988487495,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004826955104447546,
      "loss": 5.0939,
      "mean_token_accuracy": 0.17867274582386017,
      "num_tokens": 29672231.0,
      "step": 13535
    },
    {
      "entropy": 5.536306858062744,
      "epoch": 1.3437872171496625,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004826818815816114,
      "loss": 5.1445,
      "mean_token_accuracy": 0.17367480844259262,
      "num_tokens": 29683019.0,
      "step": 13540
    },
    {
      "entropy": 5.534232902526855,
      "epoch": 1.34428344581183,
      "grad_norm": 1.0,
      "learning_rate": 0.00048266824756840513,
      "loss": 5.1594,
      "mean_token_accuracy": 0.17224482148885728,
      "num_tokens": 29694233.0,
      "step": 13545
    },
    {
      "entropy": 5.562705039978027,
      "epoch": 1.3447796744739975,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004826546084054737,
      "loss": 5.1977,
      "mean_token_accuracy": 0.16854349821805953,
      "num_tokens": 29704023.0,
      "step": 13550
    },
    {
      "entropy": 5.641705417633057,
      "epoch": 1.3452759031361652,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004826409640931555,
      "loss": 5.2865,
      "mean_token_accuracy": 0.16259726881980896,
      "num_tokens": 29715407.0,
      "step": 13555
    },
    {
      "entropy": 5.48348593711853,
      "epoch": 1.3457721317983327,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048262731463178884,
      "loss": 5.1015,
      "mean_token_accuracy": 0.17188044041395187,
      "num_tokens": 29726223.0,
      "step": 13560
    },
    {
      "entropy": 5.495322227478027,
      "epoch": 1.3462683604605001,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004826136600217121,
      "loss": 5.0278,
      "mean_token_accuracy": 0.17780915945768355,
      "num_tokens": 29735850.0,
      "step": 13565
    },
    {
      "entropy": 5.557669258117675,
      "epoch": 1.3467645891226678,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000482600000263264,
      "loss": 5.1982,
      "mean_token_accuracy": 0.1690421849489212,
      "num_tokens": 29747381.0,
      "step": 13570
    },
    {
      "entropy": 5.56163182258606,
      "epoch": 1.3472608177848353,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004825863353567832,
      "loss": 5.2256,
      "mean_token_accuracy": 0.16919496208429335,
      "num_tokens": 29758440.0,
      "step": 13575
    },
    {
      "entropy": 5.4747802257537845,
      "epoch": 1.3477570464470028,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048257266530260865,
      "loss": 5.0804,
      "mean_token_accuracy": 0.17054817378520964,
      "num_tokens": 29770606.0,
      "step": 13580
    },
    {
      "entropy": 5.508941602706909,
      "epoch": 1.3482532751091703,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004825589901010792,
      "loss": 5.1555,
      "mean_token_accuracy": 0.17294624596834182,
      "num_tokens": 29781469.0,
      "step": 13585
    },
    {
      "entropy": 5.538801288604736,
      "epoch": 1.3487495037713377,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004825453097525341,
      "loss": 5.2181,
      "mean_token_accuracy": 0.16858044266700745,
      "num_tokens": 29792847.0,
      "step": 13590
    },
    {
      "entropy": 5.5267258644104,
      "epoch": 1.3492457324335054,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048253162425731253,
      "loss": 5.131,
      "mean_token_accuracy": 0.17018972039222718,
      "num_tokens": 29802784.0,
      "step": 13595
    },
    {
      "entropy": 5.582158184051513,
      "epoch": 1.349741961095673,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004825179336157539,
      "loss": 5.1481,
      "mean_token_accuracy": 0.17191785126924514,
      "num_tokens": 29812849.0,
      "step": 13600
    },
    {
      "entropy": 5.526828813552856,
      "epoch": 1.3502381897578404,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048250423782819765,
      "loss": 5.1204,
      "mean_token_accuracy": 0.16780119240283967,
      "num_tokens": 29823495.0,
      "step": 13605
    },
    {
      "entropy": 5.529862880706787,
      "epoch": 1.350734418420008,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004824905368949834,
      "loss": 5.1778,
      "mean_token_accuracy": 0.1669362425804138,
      "num_tokens": 29835295.0,
      "step": 13610
    },
    {
      "entropy": 5.510252904891968,
      "epoch": 1.3512306470821756,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004824768308164509,
      "loss": 5.1085,
      "mean_token_accuracy": 0.176770481467247,
      "num_tokens": 29845747.0,
      "step": 13615
    },
    {
      "entropy": 5.538020896911621,
      "epoch": 1.351726875744343,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004824631195929402,
      "loss": 5.171,
      "mean_token_accuracy": 0.17418198883533478,
      "num_tokens": 29857111.0,
      "step": 13620
    },
    {
      "entropy": 5.494154548645019,
      "epoch": 1.3522231044065105,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004824494032247911,
      "loss": 5.1597,
      "mean_token_accuracy": 0.164189974963665,
      "num_tokens": 29866919.0,
      "step": 13625
    },
    {
      "entropy": 5.511776733398437,
      "epoch": 1.352719333068678,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004824356817123439,
      "loss": 5.1129,
      "mean_token_accuracy": 0.16926506757736207,
      "num_tokens": 29878631.0,
      "step": 13630
    },
    {
      "entropy": 5.524524545669555,
      "epoch": 1.3532155617308455,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048242195505593865,
      "loss": 5.1422,
      "mean_token_accuracy": 0.17134637683629989,
      "num_tokens": 29890809.0,
      "step": 13635
    },
    {
      "entropy": 5.562575626373291,
      "epoch": 1.3537117903930131,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048240822325591596,
      "loss": 5.2187,
      "mean_token_accuracy": 0.17203592360019684,
      "num_tokens": 29901789.0,
      "step": 13640
    },
    {
      "entropy": 5.498739385604859,
      "epoch": 1.3542080190551806,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004823944863126162,
      "loss": 5.1246,
      "mean_token_accuracy": 0.16743547171354295,
      "num_tokens": 29912576.0,
      "step": 13645
    },
    {
      "entropy": 5.580034875869751,
      "epoch": 1.354704247717348,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004823807442263801,
      "loss": 5.2168,
      "mean_token_accuracy": 0.1606443256139755,
      "num_tokens": 29924076.0,
      "step": 13650
    },
    {
      "entropy": 5.504463481903076,
      "epoch": 1.3552004763795158,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048236699699754845,
      "loss": 5.1359,
      "mean_token_accuracy": 0.1691555291414261,
      "num_tokens": 29935854.0,
      "step": 13655
    },
    {
      "entropy": 5.536391353607177,
      "epoch": 1.3556967050416833,
      "grad_norm": 0.875,
      "learning_rate": 0.000482353244626462,
      "loss": 5.093,
      "mean_token_accuracy": 0.1683717966079712,
      "num_tokens": 29946389.0,
      "step": 13660
    },
    {
      "entropy": 5.559522819519043,
      "epoch": 1.3561929337038507,
      "grad_norm": 1.0,
      "learning_rate": 0.00048233948711346196,
      "loss": 5.1595,
      "mean_token_accuracy": 0.16595248281955718,
      "num_tokens": 29958170.0,
      "step": 13665
    },
    {
      "entropy": 5.485475683212281,
      "epoch": 1.3566891623660182,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048232572445888945,
      "loss": 5.0464,
      "mean_token_accuracy": 0.1755625918507576,
      "num_tokens": 29969067.0,
      "step": 13670
    },
    {
      "entropy": 5.565349864959717,
      "epoch": 1.3571853910281857,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004823119566630857,
      "loss": 5.1858,
      "mean_token_accuracy": 0.16782769113779067,
      "num_tokens": 29979727.0,
      "step": 13675
    },
    {
      "entropy": 5.573714685440064,
      "epoch": 1.3576816196903534,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048229818372639216,
      "loss": 5.1516,
      "mean_token_accuracy": 0.17022989243268966,
      "num_tokens": 29989397.0,
      "step": 13680
    },
    {
      "entropy": 5.467497873306274,
      "epoch": 1.3581778483525209,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048228440564915026,
      "loss": 5.0803,
      "mean_token_accuracy": 0.1684701457619667,
      "num_tokens": 29999415.0,
      "step": 13685
    },
    {
      "entropy": 5.51286563873291,
      "epoch": 1.3586740770146883,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048227062243170186,
      "loss": 5.1199,
      "mean_token_accuracy": 0.17517461329698564,
      "num_tokens": 30010277.0,
      "step": 13690
    },
    {
      "entropy": 5.466383600234986,
      "epoch": 1.3591703056768558,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004822568340743886,
      "loss": 5.0119,
      "mean_token_accuracy": 0.18072010576725006,
      "num_tokens": 30020684.0,
      "step": 13695
    },
    {
      "entropy": 5.600068140029907,
      "epoch": 1.3596665343390235,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004822430405775525,
      "loss": 5.2303,
      "mean_token_accuracy": 0.16519418954849244,
      "num_tokens": 30030916.0,
      "step": 13700
    },
    {
      "entropy": 5.595063591003418,
      "epoch": 1.360162763001191,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004822292419415355,
      "loss": 5.1497,
      "mean_token_accuracy": 0.16804620772600173,
      "num_tokens": 30041140.0,
      "step": 13705
    },
    {
      "entropy": 5.51181640625,
      "epoch": 1.3606589916633585,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048221543816667984,
      "loss": 5.1126,
      "mean_token_accuracy": 0.18026674091815947,
      "num_tokens": 30052765.0,
      "step": 13710
    },
    {
      "entropy": 5.4777802467346195,
      "epoch": 1.361155220325526,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048220162925332786,
      "loss": 5.1272,
      "mean_token_accuracy": 0.1705651581287384,
      "num_tokens": 30062511.0,
      "step": 13715
    },
    {
      "entropy": 5.622525358200074,
      "epoch": 1.3616514489876934,
      "grad_norm": 1.296875,
      "learning_rate": 0.000482187815201822,
      "loss": 5.2131,
      "mean_token_accuracy": 0.16916419714689254,
      "num_tokens": 30073404.0,
      "step": 13720
    },
    {
      "entropy": 5.506596279144287,
      "epoch": 1.362147677649861,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048217399601250464,
      "loss": 5.0694,
      "mean_token_accuracy": 0.17430136799812318,
      "num_tokens": 30084978.0,
      "step": 13725
    },
    {
      "entropy": 5.538654232025147,
      "epoch": 1.3626439063120286,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004821601716857187,
      "loss": 5.1431,
      "mean_token_accuracy": 0.17120698988437652,
      "num_tokens": 30094498.0,
      "step": 13730
    },
    {
      "entropy": 5.496025657653808,
      "epoch": 1.363140134974196,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048214634222180686,
      "loss": 5.1591,
      "mean_token_accuracy": 0.17425190806388854,
      "num_tokens": 30104545.0,
      "step": 13735
    },
    {
      "entropy": 5.588847208023071,
      "epoch": 1.3636363636363638,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004821325076211121,
      "loss": 5.2407,
      "mean_token_accuracy": 0.17144907712936402,
      "num_tokens": 30114863.0,
      "step": 13740
    },
    {
      "entropy": 5.591871690750122,
      "epoch": 1.3641325922985312,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00048211866788397753,
      "loss": 5.1412,
      "mean_token_accuracy": 0.1668848067522049,
      "num_tokens": 30125101.0,
      "step": 13745
    },
    {
      "entropy": 5.5326512336730955,
      "epoch": 1.3646288209606987,
      "grad_norm": 1.125,
      "learning_rate": 0.0004821048230107462,
      "loss": 5.2082,
      "mean_token_accuracy": 0.1738513931632042,
      "num_tokens": 30135081.0,
      "step": 13750
    },
    {
      "entropy": 5.6686928272247314,
      "epoch": 1.3651250496228662,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004820909730017616,
      "loss": 5.2494,
      "mean_token_accuracy": 0.1575665593147278,
      "num_tokens": 30147169.0,
      "step": 13755
    },
    {
      "entropy": 5.563359212875366,
      "epoch": 1.3656212782850337,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048207711785736707,
      "loss": 5.0489,
      "mean_token_accuracy": 0.17081768214702606,
      "num_tokens": 30157523.0,
      "step": 13760
    },
    {
      "entropy": 5.461745977401733,
      "epoch": 1.3661175069472011,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048206325757790624,
      "loss": 5.1337,
      "mean_token_accuracy": 0.17506172955036164,
      "num_tokens": 30168313.0,
      "step": 13765
    },
    {
      "entropy": 5.461941576004028,
      "epoch": 1.3666137356093688,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048204939216372285,
      "loss": 5.0874,
      "mean_token_accuracy": 0.1732782557606697,
      "num_tokens": 30179173.0,
      "step": 13770
    },
    {
      "entropy": 5.602393531799317,
      "epoch": 1.3671099642715363,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004820355216151607,
      "loss": 5.1646,
      "mean_token_accuracy": 0.16877421587705613,
      "num_tokens": 30191666.0,
      "step": 13775
    },
    {
      "entropy": 5.600204801559448,
      "epoch": 1.3676061929337038,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048202164593256363,
      "loss": 5.1257,
      "mean_token_accuracy": 0.17001240849494934,
      "num_tokens": 30202752.0,
      "step": 13780
    },
    {
      "entropy": 5.517510795593262,
      "epoch": 1.3681024215958715,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048200776511627585,
      "loss": 5.1529,
      "mean_token_accuracy": 0.1731338307261467,
      "num_tokens": 30213127.0,
      "step": 13785
    },
    {
      "entropy": 5.57043423652649,
      "epoch": 1.368598650258039,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004819938791666416,
      "loss": 5.191,
      "mean_token_accuracy": 0.17228126674890518,
      "num_tokens": 30224782.0,
      "step": 13790
    },
    {
      "entropy": 5.538125705718994,
      "epoch": 1.3690948789202064,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048197998808400524,
      "loss": 5.1864,
      "mean_token_accuracy": 0.1719948098063469,
      "num_tokens": 30235129.0,
      "step": 13795
    },
    {
      "entropy": 5.557019567489624,
      "epoch": 1.369591107582374,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004819660918687111,
      "loss": 5.2072,
      "mean_token_accuracy": 0.16350836157798768,
      "num_tokens": 30247304.0,
      "step": 13800
    },
    {
      "entropy": 5.519007587432862,
      "epoch": 1.3700873362445414,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004819521905211039,
      "loss": 5.0757,
      "mean_token_accuracy": 0.17202291637659073,
      "num_tokens": 30257900.0,
      "step": 13805
    },
    {
      "entropy": 5.616189479827881,
      "epoch": 1.370583564906709,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048193828404152826,
      "loss": 5.2009,
      "mean_token_accuracy": 0.16968987584114076,
      "num_tokens": 30269741.0,
      "step": 13810
    },
    {
      "entropy": 5.531930732727051,
      "epoch": 1.3710797935688765,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004819243724303291,
      "loss": 5.2029,
      "mean_token_accuracy": 0.16319651156663895,
      "num_tokens": 30279382.0,
      "step": 13815
    },
    {
      "entropy": 5.553898525238037,
      "epoch": 1.371576022231044,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048191045568785144,
      "loss": 5.175,
      "mean_token_accuracy": 0.1693979024887085,
      "num_tokens": 30289837.0,
      "step": 13820
    },
    {
      "entropy": 5.514196348190308,
      "epoch": 1.3720722508932117,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048189653381444024,
      "loss": 5.1263,
      "mean_token_accuracy": 0.17463000416755675,
      "num_tokens": 30302138.0,
      "step": 13825
    },
    {
      "entropy": 5.561945962905884,
      "epoch": 1.3725684795553792,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004818826068104409,
      "loss": 5.2059,
      "mean_token_accuracy": 0.1698312595486641,
      "num_tokens": 30313623.0,
      "step": 13830
    },
    {
      "entropy": 5.574228954315186,
      "epoch": 1.3730647082175467,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004818686746761987,
      "loss": 5.1538,
      "mean_token_accuracy": 0.1668202981352806,
      "num_tokens": 30324533.0,
      "step": 13835
    },
    {
      "entropy": 5.5970619201660154,
      "epoch": 1.3735609368797141,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004818547374120591,
      "loss": 5.2531,
      "mean_token_accuracy": 0.15952227115631104,
      "num_tokens": 30335819.0,
      "step": 13840
    },
    {
      "entropy": 5.530732536315918,
      "epoch": 1.3740571655418816,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004818407950183678,
      "loss": 5.1269,
      "mean_token_accuracy": 0.17419376373291015,
      "num_tokens": 30345947.0,
      "step": 13845
    },
    {
      "entropy": 5.502305793762207,
      "epoch": 1.374553394204049,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004818268474954704,
      "loss": 5.1146,
      "mean_token_accuracy": 0.17125113159418107,
      "num_tokens": 30356972.0,
      "step": 13850
    },
    {
      "entropy": 5.584651374816895,
      "epoch": 1.3750496228662168,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004818128948437129,
      "loss": 5.1962,
      "mean_token_accuracy": 0.16647759228944778,
      "num_tokens": 30368414.0,
      "step": 13855
    },
    {
      "entropy": 5.595959424972534,
      "epoch": 1.3755458515283843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004817989370634412,
      "loss": 5.1248,
      "mean_token_accuracy": 0.17206144481897354,
      "num_tokens": 30379141.0,
      "step": 13860
    },
    {
      "entropy": 5.536151313781739,
      "epoch": 1.3760420801905517,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048178497415500153,
      "loss": 5.0588,
      "mean_token_accuracy": 0.18021752834320068,
      "num_tokens": 30388985.0,
      "step": 13865
    },
    {
      "entropy": 5.529819536209106,
      "epoch": 1.3765383088527194,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004817710061187401,
      "loss": 5.1155,
      "mean_token_accuracy": 0.18031678348779678,
      "num_tokens": 30399599.0,
      "step": 13870
    },
    {
      "entropy": 5.560130500793457,
      "epoch": 1.377034537514887,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048175703295500323,
      "loss": 5.2095,
      "mean_token_accuracy": 0.1690106675028801,
      "num_tokens": 30411143.0,
      "step": 13875
    },
    {
      "entropy": 5.634877109527588,
      "epoch": 1.3775307661770544,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048174305466413755,
      "loss": 5.1973,
      "mean_token_accuracy": 0.17103828340768815,
      "num_tokens": 30422339.0,
      "step": 13880
    },
    {
      "entropy": 5.531447649002075,
      "epoch": 1.3780269948392219,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048172907124648947,
      "loss": 5.1205,
      "mean_token_accuracy": 0.17195411920547485,
      "num_tokens": 30432588.0,
      "step": 13885
    },
    {
      "entropy": 5.566704368591308,
      "epoch": 1.3785232235013893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004817150827024059,
      "loss": 5.1805,
      "mean_token_accuracy": 0.17412401586771012,
      "num_tokens": 30442272.0,
      "step": 13890
    },
    {
      "entropy": 5.504783725738525,
      "epoch": 1.379019452163557,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004817010890322338,
      "loss": 5.0915,
      "mean_token_accuracy": 0.17417070567607879,
      "num_tokens": 30453142.0,
      "step": 13895
    },
    {
      "entropy": 5.577582311630249,
      "epoch": 1.3795156808257245,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048168709023632006,
      "loss": 5.2018,
      "mean_token_accuracy": 0.16694744378328324,
      "num_tokens": 30465136.0,
      "step": 13900
    },
    {
      "entropy": 5.635681009292602,
      "epoch": 1.380011909487892,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048167308631501184,
      "loss": 5.1228,
      "mean_token_accuracy": 0.17227154970169067,
      "num_tokens": 30476511.0,
      "step": 13905
    },
    {
      "entropy": 5.577196025848389,
      "epoch": 1.3805081381500597,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048165907726865643,
      "loss": 5.2315,
      "mean_token_accuracy": 0.1627224341034889,
      "num_tokens": 30486547.0,
      "step": 13910
    },
    {
      "entropy": 5.573952627182007,
      "epoch": 1.3810043668122272,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004816450630976012,
      "loss": 5.1948,
      "mean_token_accuracy": 0.17420513927936554,
      "num_tokens": 30496163.0,
      "step": 13915
    },
    {
      "entropy": 5.555355358123779,
      "epoch": 1.3815005954743946,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004816310438021937,
      "loss": 5.1876,
      "mean_token_accuracy": 0.17177563160657883,
      "num_tokens": 30506979.0,
      "step": 13920
    },
    {
      "entropy": 5.58404278755188,
      "epoch": 1.381996824136562,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048161701938278156,
      "loss": 5.2208,
      "mean_token_accuracy": 0.1606489047408104,
      "num_tokens": 30517492.0,
      "step": 13925
    },
    {
      "entropy": 5.500133323669433,
      "epoch": 1.3824930527987296,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048160298983971256,
      "loss": 5.0776,
      "mean_token_accuracy": 0.1745465651154518,
      "num_tokens": 30528614.0,
      "step": 13930
    },
    {
      "entropy": 5.528719902038574,
      "epoch": 1.382989281460897,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004815889551733346,
      "loss": 5.1855,
      "mean_token_accuracy": 0.16944511532783507,
      "num_tokens": 30539479.0,
      "step": 13935
    },
    {
      "entropy": 5.614548969268799,
      "epoch": 1.3834855101230648,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048157491538399573,
      "loss": 5.2142,
      "mean_token_accuracy": 0.16844443678855897,
      "num_tokens": 30549909.0,
      "step": 13940
    },
    {
      "entropy": 5.598271369934082,
      "epoch": 1.3839817387852322,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048156087047204407,
      "loss": 5.0996,
      "mean_token_accuracy": 0.1766225203871727,
      "num_tokens": 30562265.0,
      "step": 13945
    },
    {
      "entropy": 5.582998132705688,
      "epoch": 1.3844779674473997,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004815468204378279,
      "loss": 5.1352,
      "mean_token_accuracy": 0.16981200724840165,
      "num_tokens": 30572622.0,
      "step": 13950
    },
    {
      "entropy": 5.509753274917602,
      "epoch": 1.3849741961095674,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004815327652816957,
      "loss": 5.1119,
      "mean_token_accuracy": 0.17214690297842025,
      "num_tokens": 30583300.0,
      "step": 13955
    },
    {
      "entropy": 5.514261627197266,
      "epoch": 1.3854704247717349,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004815187050039959,
      "loss": 5.0686,
      "mean_token_accuracy": 0.17511713355779648,
      "num_tokens": 30594460.0,
      "step": 13960
    },
    {
      "entropy": 5.537472295761108,
      "epoch": 1.3859666534339024,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048150463960507724,
      "loss": 5.1024,
      "mean_token_accuracy": 0.1711593374609947,
      "num_tokens": 30604658.0,
      "step": 13965
    },
    {
      "entropy": 5.547265911102295,
      "epoch": 1.3864628820960698,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048149056908528855,
      "loss": 5.1047,
      "mean_token_accuracy": 0.16818841993808747,
      "num_tokens": 30616151.0,
      "step": 13970
    },
    {
      "entropy": 5.531367063522339,
      "epoch": 1.3869591107582373,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004814764934449786,
      "loss": 5.1633,
      "mean_token_accuracy": 0.1750291481614113,
      "num_tokens": 30627560.0,
      "step": 13975
    },
    {
      "entropy": 5.536222124099732,
      "epoch": 1.387455339420405,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048146241268449665,
      "loss": 5.1423,
      "mean_token_accuracy": 0.17360917925834657,
      "num_tokens": 30638490.0,
      "step": 13980
    },
    {
      "entropy": 5.598623752593994,
      "epoch": 1.3879515680825725,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004814483268041917,
      "loss": 5.2527,
      "mean_token_accuracy": 0.1691885322332382,
      "num_tokens": 30649501.0,
      "step": 13985
    },
    {
      "entropy": 5.567948818206787,
      "epoch": 1.38844779674474,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004814342358044131,
      "loss": 5.1279,
      "mean_token_accuracy": 0.17392905205488204,
      "num_tokens": 30660479.0,
      "step": 13990
    },
    {
      "entropy": 5.64867262840271,
      "epoch": 1.3889440254069074,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048142013968551026,
      "loss": 5.3088,
      "mean_token_accuracy": 0.15852625221014022,
      "num_tokens": 30672015.0,
      "step": 13995
    },
    {
      "entropy": 5.467982387542724,
      "epoch": 1.3894402540690751,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048140603844783273,
      "loss": 5.1537,
      "mean_token_accuracy": 0.16933163851499558,
      "num_tokens": 30682612.0,
      "step": 14000
    },
    {
      "entropy": 5.495663833618164,
      "epoch": 1.3899364827312426,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004813919320917303,
      "loss": 5.0957,
      "mean_token_accuracy": 0.1712799847126007,
      "num_tokens": 30693289.0,
      "step": 14005
    },
    {
      "entropy": 5.572775840759277,
      "epoch": 1.39043271139341,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048137782061755265,
      "loss": 5.1474,
      "mean_token_accuracy": 0.16950712203979493,
      "num_tokens": 30705346.0,
      "step": 14010
    },
    {
      "entropy": 5.509809923171997,
      "epoch": 1.3909289400555775,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048136370402564974,
      "loss": 5.1506,
      "mean_token_accuracy": 0.1659905955195427,
      "num_tokens": 30715968.0,
      "step": 14015
    },
    {
      "entropy": 5.519743728637695,
      "epoch": 1.391425168717745,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004813495823163716,
      "loss": 5.0999,
      "mean_token_accuracy": 0.17194507420063018,
      "num_tokens": 30726225.0,
      "step": 14020
    },
    {
      "entropy": 5.507151126861572,
      "epoch": 1.3919213973799127,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004813354554900685,
      "loss": 5.0673,
      "mean_token_accuracy": 0.1738160088658333,
      "num_tokens": 30735603.0,
      "step": 14025
    },
    {
      "entropy": 5.487907886505127,
      "epoch": 1.3924176260420802,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004813213235470907,
      "loss": 5.1307,
      "mean_token_accuracy": 0.16985554099082947,
      "num_tokens": 30747414.0,
      "step": 14030
    },
    {
      "entropy": 5.582440805435181,
      "epoch": 1.3929138547042477,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004813071864877887,
      "loss": 5.1873,
      "mean_token_accuracy": 0.17258260846138002,
      "num_tokens": 30758389.0,
      "step": 14035
    },
    {
      "entropy": 5.588983154296875,
      "epoch": 1.3934100833664154,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000481293044312513,
      "loss": 5.203,
      "mean_token_accuracy": 0.16748234033584594,
      "num_tokens": 30769484.0,
      "step": 14040
    },
    {
      "entropy": 5.490986490249634,
      "epoch": 1.3939063120285828,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048127889702161435,
      "loss": 5.1042,
      "mean_token_accuracy": 0.17436276972293854,
      "num_tokens": 30779775.0,
      "step": 14045
    },
    {
      "entropy": 5.558767890930175,
      "epoch": 1.3944025406907503,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004812647446154435,
      "loss": 5.2127,
      "mean_token_accuracy": 0.1640891522169113,
      "num_tokens": 30791456.0,
      "step": 14050
    },
    {
      "entropy": 5.537540006637573,
      "epoch": 1.3948987693529178,
      "grad_norm": 0.875,
      "learning_rate": 0.0004812505870943515,
      "loss": 5.0344,
      "mean_token_accuracy": 0.17639026194810867,
      "num_tokens": 30801659.0,
      "step": 14055
    },
    {
      "entropy": 5.624769496917724,
      "epoch": 1.3953949980150853,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004812364244586893,
      "loss": 5.2659,
      "mean_token_accuracy": 0.16538924127817153,
      "num_tokens": 30813389.0,
      "step": 14060
    },
    {
      "entropy": 5.477582693099976,
      "epoch": 1.395891226677253,
      "grad_norm": 1.0,
      "learning_rate": 0.0004812222567088082,
      "loss": 5.0251,
      "mean_token_accuracy": 0.1695793554186821,
      "num_tokens": 30823725.0,
      "step": 14065
    },
    {
      "entropy": 5.483001232147217,
      "epoch": 1.3963874553394204,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004812080838450596,
      "loss": 5.0484,
      "mean_token_accuracy": 0.17452310621738434,
      "num_tokens": 30834448.0,
      "step": 14070
    },
    {
      "entropy": 5.534590435028076,
      "epoch": 1.396883684001588,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004811939058677947,
      "loss": 5.0952,
      "mean_token_accuracy": 0.17181549817323685,
      "num_tokens": 30845891.0,
      "step": 14075
    },
    {
      "entropy": 5.519650173187256,
      "epoch": 1.3973799126637554,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004811797227773654,
      "loss": 5.0451,
      "mean_token_accuracy": 0.1788552522659302,
      "num_tokens": 30855170.0,
      "step": 14080
    },
    {
      "entropy": 5.499523878097534,
      "epoch": 1.397876141325923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048116553457412317,
      "loss": 5.1354,
      "mean_token_accuracy": 0.17040596008300782,
      "num_tokens": 30866527.0,
      "step": 14085
    },
    {
      "entropy": 5.500823593139648,
      "epoch": 1.3983723699880906,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048115134125841993,
      "loss": 5.0865,
      "mean_token_accuracy": 0.17504869252443314,
      "num_tokens": 30876292.0,
      "step": 14090
    },
    {
      "entropy": 5.5390700340271,
      "epoch": 1.398868598650258,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004811371428306076,
      "loss": 5.0694,
      "mean_token_accuracy": 0.1723286345601082,
      "num_tokens": 30887440.0,
      "step": 14095
    },
    {
      "entropy": 5.523568105697632,
      "epoch": 1.3993648273124255,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004811229392910384,
      "loss": 5.1191,
      "mean_token_accuracy": 0.17340546995401382,
      "num_tokens": 30897925.0,
      "step": 14100
    },
    {
      "entropy": 5.484356164932251,
      "epoch": 1.399861055974593,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004811087306400644,
      "loss": 5.1529,
      "mean_token_accuracy": 0.17690719962120055,
      "num_tokens": 30908413.0,
      "step": 14105
    },
    {
      "entropy": 5.580122327804565,
      "epoch": 1.4003572846367607,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048109451687803804,
      "loss": 5.1758,
      "mean_token_accuracy": 0.16955263614654542,
      "num_tokens": 30919149.0,
      "step": 14110
    },
    {
      "entropy": 5.6071288108825685,
      "epoch": 1.4008535132989282,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004810802980053117,
      "loss": 5.2003,
      "mean_token_accuracy": 0.16742929220199584,
      "num_tokens": 30929883.0,
      "step": 14115
    },
    {
      "entropy": 5.519658041000366,
      "epoch": 1.4013497419610956,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048106607402223815,
      "loss": 5.1158,
      "mean_token_accuracy": 0.16678639501333237,
      "num_tokens": 30941358.0,
      "step": 14120
    },
    {
      "entropy": 5.548413944244385,
      "epoch": 1.4018459706232633,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048105184492916995,
      "loss": 5.1051,
      "mean_token_accuracy": 0.17210416346788407,
      "num_tokens": 30953289.0,
      "step": 14125
    },
    {
      "entropy": 5.518923664093018,
      "epoch": 1.4023421992854308,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048103761072645993,
      "loss": 5.0465,
      "mean_token_accuracy": 0.17729029059410095,
      "num_tokens": 30964304.0,
      "step": 14130
    },
    {
      "entropy": 5.471157598495483,
      "epoch": 1.4028384279475983,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048102337141446116,
      "loss": 5.0713,
      "mean_token_accuracy": 0.1787961408495903,
      "num_tokens": 30974840.0,
      "step": 14135
    },
    {
      "entropy": 5.501088714599609,
      "epoch": 1.4033346566097658,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048100912699352666,
      "loss": 5.131,
      "mean_token_accuracy": 0.17864849269390107,
      "num_tokens": 30985805.0,
      "step": 14140
    },
    {
      "entropy": 5.557447481155395,
      "epoch": 1.4038308852719332,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004809948774640098,
      "loss": 5.1365,
      "mean_token_accuracy": 0.1727934792637825,
      "num_tokens": 30998174.0,
      "step": 14145
    },
    {
      "entropy": 5.559416151046753,
      "epoch": 1.4043271139341007,
      "grad_norm": 0.875,
      "learning_rate": 0.0004809806228262639,
      "loss": 5.1466,
      "mean_token_accuracy": 0.17206069678068162,
      "num_tokens": 31010050.0,
      "step": 14150
    },
    {
      "entropy": 5.612583446502685,
      "epoch": 1.4048233425962684,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048096636308064236,
      "loss": 5.2717,
      "mean_token_accuracy": 0.16032503694295883,
      "num_tokens": 31023069.0,
      "step": 14155
    },
    {
      "entropy": 5.556841230392456,
      "epoch": 1.4053195712584359,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004809520982274989,
      "loss": 5.0798,
      "mean_token_accuracy": 0.17819133996963502,
      "num_tokens": 31033512.0,
      "step": 14160
    },
    {
      "entropy": 5.515301084518432,
      "epoch": 1.4058157999206033,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048093782826718706,
      "loss": 5.1136,
      "mean_token_accuracy": 0.16862879693508148,
      "num_tokens": 31045731.0,
      "step": 14165
    },
    {
      "entropy": 5.493713808059693,
      "epoch": 1.406312028582771,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048092355320006086,
      "loss": 5.0293,
      "mean_token_accuracy": 0.1763712137937546,
      "num_tokens": 31055443.0,
      "step": 14170
    },
    {
      "entropy": 5.48050217628479,
      "epoch": 1.4068082572449385,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004809092730264744,
      "loss": 5.1242,
      "mean_token_accuracy": 0.17357977628707885,
      "num_tokens": 31067575.0,
      "step": 14175
    },
    {
      "entropy": 5.529447078704834,
      "epoch": 1.407304485907106,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048089498774678155,
      "loss": 5.2136,
      "mean_token_accuracy": 0.16673895865678787,
      "num_tokens": 31078198.0,
      "step": 14180
    },
    {
      "entropy": 5.519463539123535,
      "epoch": 1.4078007145692735,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048088069736133664,
      "loss": 5.0996,
      "mean_token_accuracy": 0.17052169889211655,
      "num_tokens": 31089961.0,
      "step": 14185
    },
    {
      "entropy": 5.537969446182251,
      "epoch": 1.408296943231441,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004808664018704941,
      "loss": 5.1126,
      "mean_token_accuracy": 0.1709460660815239,
      "num_tokens": 31100534.0,
      "step": 14190
    },
    {
      "entropy": 5.641386461257935,
      "epoch": 1.4087931718936086,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004808521012746085,
      "loss": 5.1461,
      "mean_token_accuracy": 0.17591289877891542,
      "num_tokens": 31112016.0,
      "step": 14195
    },
    {
      "entropy": 5.498534440994263,
      "epoch": 1.4092894005557761,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048083779557403424,
      "loss": 5.1653,
      "mean_token_accuracy": 0.17351087778806687,
      "num_tokens": 31123659.0,
      "step": 14200
    },
    {
      "entropy": 5.54283504486084,
      "epoch": 1.4097856292179436,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004808234847691262,
      "loss": 5.1138,
      "mean_token_accuracy": 0.16809820830821992,
      "num_tokens": 31133872.0,
      "step": 14205
    },
    {
      "entropy": 5.57604603767395,
      "epoch": 1.4102818578801113,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004808091688602393,
      "loss": 5.1701,
      "mean_token_accuracy": 0.16956085115671157,
      "num_tokens": 31145834.0,
      "step": 14210
    },
    {
      "entropy": 5.564661598205566,
      "epoch": 1.4107780865422788,
      "grad_norm": 0.875,
      "learning_rate": 0.00048079484784772854,
      "loss": 5.2306,
      "mean_token_accuracy": 0.16539136320352554,
      "num_tokens": 31157037.0,
      "step": 14215
    },
    {
      "entropy": 5.532263612747192,
      "epoch": 1.4112743152044462,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048078052173194886,
      "loss": 5.1133,
      "mean_token_accuracy": 0.17727053314447402,
      "num_tokens": 31166812.0,
      "step": 14220
    },
    {
      "entropy": 5.548606300354004,
      "epoch": 1.4117705438666137,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004807661905132557,
      "loss": 5.1133,
      "mean_token_accuracy": 0.17000300586223602,
      "num_tokens": 31177162.0,
      "step": 14225
    },
    {
      "entropy": 5.551578426361084,
      "epoch": 1.4122667725287812,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004807518541920045,
      "loss": 5.1521,
      "mean_token_accuracy": 0.17382025569677353,
      "num_tokens": 31187333.0,
      "step": 14230
    },
    {
      "entropy": 5.581780099868775,
      "epoch": 1.4127630011909487,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048073751276855056,
      "loss": 5.1608,
      "mean_token_accuracy": 0.16805825978517533,
      "num_tokens": 31197869.0,
      "step": 14235
    },
    {
      "entropy": 5.555130338668823,
      "epoch": 1.4132592298531164,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004807231662432497,
      "loss": 5.164,
      "mean_token_accuracy": 0.1665928155183792,
      "num_tokens": 31208199.0,
      "step": 14240
    },
    {
      "entropy": 5.557612657546997,
      "epoch": 1.4137554585152838,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004807088146164575,
      "loss": 5.1467,
      "mean_token_accuracy": 0.1648218020796776,
      "num_tokens": 31218903.0,
      "step": 14245
    },
    {
      "entropy": 5.517922306060791,
      "epoch": 1.4142516871774513,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004806944578885301,
      "loss": 5.1555,
      "mean_token_accuracy": 0.1690744251012802,
      "num_tokens": 31231146.0,
      "step": 14250
    },
    {
      "entropy": 5.541270542144775,
      "epoch": 1.414747915839619,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048068009605982333,
      "loss": 5.0739,
      "mean_token_accuracy": 0.1735615149140358,
      "num_tokens": 31240980.0,
      "step": 14255
    },
    {
      "entropy": 5.5217297077178955,
      "epoch": 1.4152441445017865,
      "grad_norm": 1.0,
      "learning_rate": 0.00048066572913069334,
      "loss": 5.1142,
      "mean_token_accuracy": 0.169989150762558,
      "num_tokens": 31252458.0,
      "step": 14260
    },
    {
      "entropy": 5.514530181884766,
      "epoch": 1.415740373163954,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004806513571014965,
      "loss": 5.1039,
      "mean_token_accuracy": 0.17186874896287918,
      "num_tokens": 31262979.0,
      "step": 14265
    },
    {
      "entropy": 5.5504576683044435,
      "epoch": 1.4162366018261214,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004806369799725891,
      "loss": 5.0961,
      "mean_token_accuracy": 0.16948116570711136,
      "num_tokens": 31274811.0,
      "step": 14270
    },
    {
      "entropy": 5.507457304000854,
      "epoch": 1.416732830488289,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004806225977443277,
      "loss": 5.0625,
      "mean_token_accuracy": 0.17748325318098068,
      "num_tokens": 31285573.0,
      "step": 14275
    },
    {
      "entropy": 5.597534418106079,
      "epoch": 1.4172290591504566,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004806082104170689,
      "loss": 5.2375,
      "mean_token_accuracy": 0.16136072874069213,
      "num_tokens": 31296993.0,
      "step": 14280
    },
    {
      "entropy": 5.50703616142273,
      "epoch": 1.417725287812624,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048059381799116967,
      "loss": 5.1192,
      "mean_token_accuracy": 0.1678570181131363,
      "num_tokens": 31307217.0,
      "step": 14285
    },
    {
      "entropy": 5.491989517211914,
      "epoch": 1.4182215164747916,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004805794204669867,
      "loss": 5.1443,
      "mean_token_accuracy": 0.17037644237279892,
      "num_tokens": 31317939.0,
      "step": 14290
    },
    {
      "entropy": 5.602749443054199,
      "epoch": 1.418717745136959,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048056501784487707,
      "loss": 5.1205,
      "mean_token_accuracy": 0.16901961714029312,
      "num_tokens": 31327731.0,
      "step": 14295
    },
    {
      "entropy": 5.719781351089478,
      "epoch": 1.4192139737991267,
      "grad_norm": 1.046875,
      "learning_rate": 0.000480550610125198,
      "loss": 5.2978,
      "mean_token_accuracy": 0.15915951281785964,
      "num_tokens": 31339388.0,
      "step": 14300
    },
    {
      "entropy": 5.490464496612549,
      "epoch": 1.4197102024612942,
      "grad_norm": 1.0,
      "learning_rate": 0.0004805361973083066,
      "loss": 4.9961,
      "mean_token_accuracy": 0.17786178588867188,
      "num_tokens": 31350186.0,
      "step": 14305
    },
    {
      "entropy": 5.531382703781128,
      "epoch": 1.4202064311234617,
      "grad_norm": 1.0,
      "learning_rate": 0.0004805217793945605,
      "loss": 5.143,
      "mean_token_accuracy": 0.1653599351644516,
      "num_tokens": 31361219.0,
      "step": 14310
    },
    {
      "entropy": 5.550638246536255,
      "epoch": 1.4207026597856292,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004805073563843171,
      "loss": 5.1803,
      "mean_token_accuracy": 0.16556326001882554,
      "num_tokens": 31372820.0,
      "step": 14315
    },
    {
      "entropy": 5.536877870559692,
      "epoch": 1.4211988884477966,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048049292827793413,
      "loss": 5.0608,
      "mean_token_accuracy": 0.17393716126680375,
      "num_tokens": 31382779.0,
      "step": 14320
    },
    {
      "entropy": 5.6161552429199215,
      "epoch": 1.4216951171099643,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004804784950757693,
      "loss": 5.2638,
      "mean_token_accuracy": 0.1676454499363899,
      "num_tokens": 31394872.0,
      "step": 14325
    },
    {
      "entropy": 5.56783390045166,
      "epoch": 1.4221913457721318,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048046405677818057,
      "loss": 5.1713,
      "mean_token_accuracy": 0.17211804091930388,
      "num_tokens": 31405767.0,
      "step": 14330
    },
    {
      "entropy": 5.586855125427246,
      "epoch": 1.4226875744342993,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048044961338552595,
      "loss": 5.1709,
      "mean_token_accuracy": 0.1698562055826187,
      "num_tokens": 31417255.0,
      "step": 14335
    },
    {
      "entropy": 5.561547899246216,
      "epoch": 1.423183803096467,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004804351648981636,
      "loss": 5.1226,
      "mean_token_accuracy": 0.17501015365123748,
      "num_tokens": 31428494.0,
      "step": 14340
    },
    {
      "entropy": 5.516804885864258,
      "epoch": 1.4236800317586344,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004804207113164519,
      "loss": 5.0327,
      "mean_token_accuracy": 0.17773139029741286,
      "num_tokens": 31439221.0,
      "step": 14345
    },
    {
      "entropy": 5.537303018569946,
      "epoch": 1.424176260420802,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048040625264074915,
      "loss": 5.1175,
      "mean_token_accuracy": 0.17708641439676284,
      "num_tokens": 31450575.0,
      "step": 14350
    },
    {
      "entropy": 5.452661561965942,
      "epoch": 1.4246724890829694,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048039178887141404,
      "loss": 5.1208,
      "mean_token_accuracy": 0.17472280263900758,
      "num_tokens": 31462168.0,
      "step": 14355
    },
    {
      "entropy": 5.541571760177613,
      "epoch": 1.4251687177451369,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048037732000880503,
      "loss": 5.1417,
      "mean_token_accuracy": 0.17525738179683686,
      "num_tokens": 31472619.0,
      "step": 14360
    },
    {
      "entropy": 5.586092805862426,
      "epoch": 1.4256649464073046,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048036284605328106,
      "loss": 5.2185,
      "mean_token_accuracy": 0.1665370911359787,
      "num_tokens": 31485246.0,
      "step": 14365
    },
    {
      "entropy": 5.548651313781738,
      "epoch": 1.426161175069472,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004803483670052011,
      "loss": 5.1325,
      "mean_token_accuracy": 0.16484898328781128,
      "num_tokens": 31496258.0,
      "step": 14370
    },
    {
      "entropy": 5.525687313079834,
      "epoch": 1.4266574037316395,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000480333882864924,
      "loss": 5.1004,
      "mean_token_accuracy": 0.1732739731669426,
      "num_tokens": 31506671.0,
      "step": 14375
    },
    {
      "entropy": 5.556588077545166,
      "epoch": 1.427153632393807,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004803193936328092,
      "loss": 5.1417,
      "mean_token_accuracy": 0.166272109746933,
      "num_tokens": 31517001.0,
      "step": 14380
    },
    {
      "entropy": 5.544458532333374,
      "epoch": 1.4276498610559747,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004803048993092158,
      "loss": 5.1428,
      "mean_token_accuracy": 0.17396878749132155,
      "num_tokens": 31529649.0,
      "step": 14385
    },
    {
      "entropy": 5.462076568603516,
      "epoch": 1.4281460897181422,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048029039989450317,
      "loss": 5.0617,
      "mean_token_accuracy": 0.17557339519262313,
      "num_tokens": 31540083.0,
      "step": 14390
    },
    {
      "entropy": 5.574583101272583,
      "epoch": 1.4286423183803096,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048027589538903115,
      "loss": 5.1891,
      "mean_token_accuracy": 0.17000586837530135,
      "num_tokens": 31551400.0,
      "step": 14395
    },
    {
      "entropy": 5.558568286895752,
      "epoch": 1.4291385470424771,
      "grad_norm": 1.0,
      "learning_rate": 0.0004802613857931592,
      "loss": 5.1761,
      "mean_token_accuracy": 0.16776325404644013,
      "num_tokens": 31562430.0,
      "step": 14400
    },
    {
      "entropy": 5.561457729339599,
      "epoch": 1.4296347757046446,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004802468711072472,
      "loss": 5.2149,
      "mean_token_accuracy": 0.1718543663620949,
      "num_tokens": 31573173.0,
      "step": 14405
    },
    {
      "entropy": 5.550090551376343,
      "epoch": 1.4301310043668123,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000480232351331655,
      "loss": 5.1495,
      "mean_token_accuracy": 0.17230795919895173,
      "num_tokens": 31584072.0,
      "step": 14410
    },
    {
      "entropy": 5.483279132843018,
      "epoch": 1.4306272330289798,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048021782646674277,
      "loss": 5.0859,
      "mean_token_accuracy": 0.17675964236259462,
      "num_tokens": 31594993.0,
      "step": 14415
    },
    {
      "entropy": 5.484372758865357,
      "epoch": 1.4311234616911472,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048020329651287064,
      "loss": 5.109,
      "mean_token_accuracy": 0.17678418904542922,
      "num_tokens": 31604284.0,
      "step": 14420
    },
    {
      "entropy": 5.505160856246948,
      "epoch": 1.431619690353315,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000480188761470399,
      "loss": 5.193,
      "mean_token_accuracy": 0.16898860186338424,
      "num_tokens": 31616188.0,
      "step": 14425
    },
    {
      "entropy": 5.615882444381714,
      "epoch": 1.4321159190154824,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004801742213396881,
      "loss": 5.1558,
      "mean_token_accuracy": 0.17215863317251207,
      "num_tokens": 31626606.0,
      "step": 14430
    },
    {
      "entropy": 5.499743223190308,
      "epoch": 1.4326121476776499,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048015967612109875,
      "loss": 5.0689,
      "mean_token_accuracy": 0.18403681367635727,
      "num_tokens": 31638050.0,
      "step": 14435
    },
    {
      "entropy": 5.521497535705566,
      "epoch": 1.4331083763398174,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048014512581499136,
      "loss": 5.0896,
      "mean_token_accuracy": 0.17138998955488205,
      "num_tokens": 31648451.0,
      "step": 14440
    },
    {
      "entropy": 5.519690799713135,
      "epoch": 1.4336046050019848,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000480130570421727,
      "loss": 5.1064,
      "mean_token_accuracy": 0.16932046562433242,
      "num_tokens": 31659310.0,
      "step": 14445
    },
    {
      "entropy": 5.504786491394043,
      "epoch": 1.4341008336641523,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004801160099416665,
      "loss": 5.1378,
      "mean_token_accuracy": 0.16748925894498826,
      "num_tokens": 31669594.0,
      "step": 14450
    },
    {
      "entropy": 5.528512287139892,
      "epoch": 1.43459706232632,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004801014443751709,
      "loss": 5.14,
      "mean_token_accuracy": 0.1666400671005249,
      "num_tokens": 31682034.0,
      "step": 14455
    },
    {
      "entropy": 5.589808320999145,
      "epoch": 1.4350932909884875,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004800868737226015,
      "loss": 5.1655,
      "mean_token_accuracy": 0.17436257749795914,
      "num_tokens": 31692390.0,
      "step": 14460
    },
    {
      "entropy": 5.557921600341797,
      "epoch": 1.435589519650655,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048007229798431946,
      "loss": 5.1457,
      "mean_token_accuracy": 0.17328951209783555,
      "num_tokens": 31703690.0,
      "step": 14465
    },
    {
      "entropy": 5.545484733581543,
      "epoch": 1.4360857483128227,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004800577171606864,
      "loss": 5.1409,
      "mean_token_accuracy": 0.17311709821224214,
      "num_tokens": 31714655.0,
      "step": 14470
    },
    {
      "entropy": 5.488937044143677,
      "epoch": 1.4365819769749901,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004800431312520638,
      "loss": 5.0594,
      "mean_token_accuracy": 0.1795497789978981,
      "num_tokens": 31727022.0,
      "step": 14475
    },
    {
      "entropy": 5.548110818862915,
      "epoch": 1.4370782056371576,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004800285402588134,
      "loss": 5.1085,
      "mean_token_accuracy": 0.1772718459367752,
      "num_tokens": 31738071.0,
      "step": 14480
    },
    {
      "entropy": 5.59354567527771,
      "epoch": 1.437574434299325,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004800139441812969,
      "loss": 5.2143,
      "mean_token_accuracy": 0.16117098033428193,
      "num_tokens": 31749342.0,
      "step": 14485
    },
    {
      "entropy": 5.556843280792236,
      "epoch": 1.4380706629614926,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047999934301987644,
      "loss": 5.1927,
      "mean_token_accuracy": 0.16907583326101303,
      "num_tokens": 31761218.0,
      "step": 14490
    },
    {
      "entropy": 5.564563035964966,
      "epoch": 1.4385668916236602,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000479984736774914,
      "loss": 5.2028,
      "mean_token_accuracy": 0.16832948178052903,
      "num_tokens": 31773111.0,
      "step": 14495
    },
    {
      "entropy": 5.603694868087769,
      "epoch": 1.4390631202858277,
      "grad_norm": 1.0,
      "learning_rate": 0.00047997012544677176,
      "loss": 5.153,
      "mean_token_accuracy": 0.17313717901706696,
      "num_tokens": 31782948.0,
      "step": 14500
    },
    {
      "entropy": 5.590463399887085,
      "epoch": 1.4395593489479952,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004799555090358121,
      "loss": 5.1388,
      "mean_token_accuracy": 0.17268241345882415,
      "num_tokens": 31793992.0,
      "step": 14505
    },
    {
      "entropy": 5.566668081283569,
      "epoch": 1.440055577610163,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004799408875423975,
      "loss": 5.153,
      "mean_token_accuracy": 0.16334629505872728,
      "num_tokens": 31804830.0,
      "step": 14510
    },
    {
      "entropy": 5.515792655944824,
      "epoch": 1.4405518062723304,
      "grad_norm": 0.875,
      "learning_rate": 0.0004799262609668904,
      "loss": 5.1384,
      "mean_token_accuracy": 0.1739792615175247,
      "num_tokens": 31817200.0,
      "step": 14515
    },
    {
      "entropy": 5.52047028541565,
      "epoch": 1.4410480349344978,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004799116293096537,
      "loss": 5.1811,
      "mean_token_accuracy": 0.17165776044130326,
      "num_tokens": 31828226.0,
      "step": 14520
    },
    {
      "entropy": 5.5173523902893065,
      "epoch": 1.4415442635966653,
      "grad_norm": 1.0,
      "learning_rate": 0.0004798969925710501,
      "loss": 5.0669,
      "mean_token_accuracy": 0.18043017238378525,
      "num_tokens": 31839635.0,
      "step": 14525
    },
    {
      "entropy": 5.54227991104126,
      "epoch": 1.4420404922588328,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047988235075144254,
      "loss": 5.1097,
      "mean_token_accuracy": 0.17895270884037018,
      "num_tokens": 31850333.0,
      "step": 14530
    },
    {
      "entropy": 5.489082527160645,
      "epoch": 1.4425367209210003,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004798677038511942,
      "loss": 5.092,
      "mean_token_accuracy": 0.16857707649469375,
      "num_tokens": 31861002.0,
      "step": 14535
    },
    {
      "entropy": 5.548662710189819,
      "epoch": 1.443032949583168,
      "grad_norm": 1.0,
      "learning_rate": 0.0004798530518706683,
      "loss": 5.1678,
      "mean_token_accuracy": 0.16767743676900865,
      "num_tokens": 31871713.0,
      "step": 14540
    },
    {
      "entropy": 5.54739818572998,
      "epoch": 1.4435291782453354,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004798383948102281,
      "loss": 5.0995,
      "mean_token_accuracy": 0.16896482259035112,
      "num_tokens": 31882381.0,
      "step": 14545
    },
    {
      "entropy": 5.569896125793457,
      "epoch": 1.444025406907503,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004798237326702371,
      "loss": 5.1257,
      "mean_token_accuracy": 0.17677302062511444,
      "num_tokens": 31892347.0,
      "step": 14550
    },
    {
      "entropy": 5.626822662353516,
      "epoch": 1.4445216355696706,
      "grad_norm": 0.921875,
      "learning_rate": 0.00047980906545105884,
      "loss": 5.2405,
      "mean_token_accuracy": 0.16343855410814284,
      "num_tokens": 31902751.0,
      "step": 14555
    },
    {
      "entropy": 5.539401912689209,
      "epoch": 1.445017864231838,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004797943931530571,
      "loss": 5.1066,
      "mean_token_accuracy": 0.1770730197429657,
      "num_tokens": 31913294.0,
      "step": 14560
    },
    {
      "entropy": 5.541674518585205,
      "epoch": 1.4455140928940056,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004797797157765958,
      "loss": 5.2252,
      "mean_token_accuracy": 0.1736141175031662,
      "num_tokens": 31925678.0,
      "step": 14565
    },
    {
      "entropy": 5.477121734619141,
      "epoch": 1.446010321556173,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004797650333220387,
      "loss": 5.0455,
      "mean_token_accuracy": 0.17725862711668014,
      "num_tokens": 31934578.0,
      "step": 14570
    },
    {
      "entropy": 5.483195114135742,
      "epoch": 1.4465065502183405,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047975034578975006,
      "loss": 5.0982,
      "mean_token_accuracy": 0.17120981961488724,
      "num_tokens": 31945482.0,
      "step": 14575
    },
    {
      "entropy": 5.53853964805603,
      "epoch": 1.4470027788805082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000479735653180094,
      "loss": 5.2027,
      "mean_token_accuracy": 0.17311585694551468,
      "num_tokens": 31957303.0,
      "step": 14580
    },
    {
      "entropy": 5.544617652893066,
      "epoch": 1.4474990075426757,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004797209554934349,
      "loss": 5.1182,
      "mean_token_accuracy": 0.16760422587394713,
      "num_tokens": 31968601.0,
      "step": 14585
    },
    {
      "entropy": 5.556090831756592,
      "epoch": 1.4479952362048432,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004797062527301373,
      "loss": 5.1688,
      "mean_token_accuracy": 0.16858005821704863,
      "num_tokens": 31978987.0,
      "step": 14590
    },
    {
      "entropy": 5.530074882507324,
      "epoch": 1.4484914648670106,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004796915448905657,
      "loss": 5.1344,
      "mean_token_accuracy": 0.16919378638267518,
      "num_tokens": 31990950.0,
      "step": 14595
    },
    {
      "entropy": 5.601096820831299,
      "epoch": 1.4489876935291783,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047967683197508486,
      "loss": 5.1385,
      "mean_token_accuracy": 0.16988271921873094,
      "num_tokens": 32001795.0,
      "step": 14600
    },
    {
      "entropy": 5.574069786071777,
      "epoch": 1.4494839221913458,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047966211398405966,
      "loss": 5.1625,
      "mean_token_accuracy": 0.16445329189300537,
      "num_tokens": 32011775.0,
      "step": 14605
    },
    {
      "entropy": 5.645728683471679,
      "epoch": 1.4499801508535133,
      "grad_norm": 1.015625,
      "learning_rate": 0.000479647390917855,
      "loss": 5.2472,
      "mean_token_accuracy": 0.15949815660715103,
      "num_tokens": 32022674.0,
      "step": 14610
    },
    {
      "entropy": 5.586968660354614,
      "epoch": 1.4504763795156808,
      "grad_norm": 1.015625,
      "learning_rate": 0.000479632662776836,
      "loss": 5.202,
      "mean_token_accuracy": 0.17336668223142623,
      "num_tokens": 32033383.0,
      "step": 14615
    },
    {
      "entropy": 5.559762620925904,
      "epoch": 1.4509726081778482,
      "grad_norm": 0.953125,
      "learning_rate": 0.000479617929561368,
      "loss": 5.0982,
      "mean_token_accuracy": 0.17610179781913757,
      "num_tokens": 32044066.0,
      "step": 14620
    },
    {
      "entropy": 5.544317197799683,
      "epoch": 1.451468836840016,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004796031912718162,
      "loss": 5.1845,
      "mean_token_accuracy": 0.1644149601459503,
      "num_tokens": 32054650.0,
      "step": 14625
    },
    {
      "entropy": 5.616777038574218,
      "epoch": 1.4519650655021834,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047958844790854606,
      "loss": 5.1798,
      "mean_token_accuracy": 0.16292149126529692,
      "num_tokens": 32065786.0,
      "step": 14630
    },
    {
      "entropy": 5.607652378082276,
      "epoch": 1.4524612941643509,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047957369947192337,
      "loss": 5.1222,
      "mean_token_accuracy": 0.17663830071687697,
      "num_tokens": 32076326.0,
      "step": 14635
    },
    {
      "entropy": 5.505214834213257,
      "epoch": 1.4529575228265186,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004795589459623136,
      "loss": 5.0797,
      "mean_token_accuracy": 0.17234665751457215,
      "num_tokens": 32086915.0,
      "step": 14640
    },
    {
      "entropy": 5.524808263778686,
      "epoch": 1.453453751488686,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047954418738008295,
      "loss": 5.0431,
      "mean_token_accuracy": 0.17361759543418884,
      "num_tokens": 32096614.0,
      "step": 14645
    },
    {
      "entropy": 5.542260694503784,
      "epoch": 1.4539499801508535,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004795294237255971,
      "loss": 5.132,
      "mean_token_accuracy": 0.1735408589243889,
      "num_tokens": 32107721.0,
      "step": 14650
    },
    {
      "entropy": 5.5165914535522464,
      "epoch": 1.454446208813021,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047951465499922234,
      "loss": 5.1471,
      "mean_token_accuracy": 0.16828598380088805,
      "num_tokens": 32116621.0,
      "step": 14655
    },
    {
      "entropy": 5.576553440093994,
      "epoch": 1.4549424374751885,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004794998812013248,
      "loss": 5.1793,
      "mean_token_accuracy": 0.16863167881965638,
      "num_tokens": 32126536.0,
      "step": 14660
    },
    {
      "entropy": 5.586959505081177,
      "epoch": 1.4554386661373562,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047948510233227084,
      "loss": 5.2409,
      "mean_token_accuracy": 0.16116394251585006,
      "num_tokens": 32137059.0,
      "step": 14665
    },
    {
      "entropy": 5.518677139282227,
      "epoch": 1.4559348947995236,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004794703183924269,
      "loss": 5.1785,
      "mean_token_accuracy": 0.17294724136590958,
      "num_tokens": 32148186.0,
      "step": 14670
    },
    {
      "entropy": 5.614913368225098,
      "epoch": 1.4564311234616911,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004794555293821597,
      "loss": 5.2262,
      "mean_token_accuracy": 0.15936787873506547,
      "num_tokens": 32158244.0,
      "step": 14675
    },
    {
      "entropy": 5.5571283340454105,
      "epoch": 1.4569273521238586,
      "grad_norm": 0.953125,
      "learning_rate": 0.000479440735301836,
      "loss": 5.0787,
      "mean_token_accuracy": 0.17744283825159074,
      "num_tokens": 32168404.0,
      "step": 14680
    },
    {
      "entropy": 5.509984922409058,
      "epoch": 1.4574235807860263,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004794259361518226,
      "loss": 5.0635,
      "mean_token_accuracy": 0.17613533586263658,
      "num_tokens": 32178744.0,
      "step": 14685
    },
    {
      "entropy": 5.513152837753296,
      "epoch": 1.4579198094481938,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047941113193248643,
      "loss": 5.1795,
      "mean_token_accuracy": 0.1680013969540596,
      "num_tokens": 32189948.0,
      "step": 14690
    },
    {
      "entropy": 5.577406024932861,
      "epoch": 1.4584160381103612,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004793963226441947,
      "loss": 5.1663,
      "mean_token_accuracy": 0.16641853302717208,
      "num_tokens": 32200979.0,
      "step": 14695
    },
    {
      "entropy": 5.536780452728271,
      "epoch": 1.4589122667725287,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047938150828731455,
      "loss": 5.1467,
      "mean_token_accuracy": 0.17314921766519548,
      "num_tokens": 32211569.0,
      "step": 14700
    },
    {
      "entropy": 5.561519908905029,
      "epoch": 1.4594084954346962,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004793666888622134,
      "loss": 5.1347,
      "mean_token_accuracy": 0.1694324180483818,
      "num_tokens": 32222429.0,
      "step": 14705
    },
    {
      "entropy": 5.571921062469483,
      "epoch": 1.459904724096864,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004793518643692587,
      "loss": 5.1893,
      "mean_token_accuracy": 0.16576670706272126,
      "num_tokens": 32234332.0,
      "step": 14710
    },
    {
      "entropy": 5.620045900344849,
      "epoch": 1.4604009527590314,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004793370348088182,
      "loss": 5.2234,
      "mean_token_accuracy": 0.17236478626728058,
      "num_tokens": 32244798.0,
      "step": 14715
    },
    {
      "entropy": 5.590567874908447,
      "epoch": 1.4608971814211988,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004793222001812595,
      "loss": 5.137,
      "mean_token_accuracy": 0.1752204939723015,
      "num_tokens": 32257746.0,
      "step": 14720
    },
    {
      "entropy": 5.577586555480957,
      "epoch": 1.4613934100833665,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047930736048695045,
      "loss": 5.2356,
      "mean_token_accuracy": 0.16509189307689667,
      "num_tokens": 32270692.0,
      "step": 14725
    },
    {
      "entropy": 5.531466436386109,
      "epoch": 1.461889638745534,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004792925157262592,
      "loss": 5.1171,
      "mean_token_accuracy": 0.17350534796714784,
      "num_tokens": 32281350.0,
      "step": 14730
    },
    {
      "entropy": 5.565432596206665,
      "epoch": 1.4623858674077015,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004792776658995537,
      "loss": 5.153,
      "mean_token_accuracy": 0.1720444455742836,
      "num_tokens": 32293182.0,
      "step": 14735
    },
    {
      "entropy": 5.546879720687866,
      "epoch": 1.462882096069869,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004792628110072023,
      "loss": 5.1268,
      "mean_token_accuracy": 0.16702627539634704,
      "num_tokens": 32304903.0,
      "step": 14740
    },
    {
      "entropy": 5.620081615447998,
      "epoch": 1.4633783247320364,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004792479510495734,
      "loss": 5.3009,
      "mean_token_accuracy": 0.1631435364484787,
      "num_tokens": 32316074.0,
      "step": 14745
    },
    {
      "entropy": 5.567771053314209,
      "epoch": 1.463874553394204,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047923308602703535,
      "loss": 5.1125,
      "mean_token_accuracy": 0.18134698122739792,
      "num_tokens": 32327594.0,
      "step": 14750
    },
    {
      "entropy": 5.53625898361206,
      "epoch": 1.4643707820563716,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047921821593995684,
      "loss": 5.1584,
      "mean_token_accuracy": 0.17243526726961136,
      "num_tokens": 32339224.0,
      "step": 14755
    },
    {
      "entropy": 5.581178379058838,
      "epoch": 1.464867010718539,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004792033407887066,
      "loss": 5.1224,
      "mean_token_accuracy": 0.16923604309558868,
      "num_tokens": 32349451.0,
      "step": 14760
    },
    {
      "entropy": 5.681001234054565,
      "epoch": 1.4653632393807066,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004791884605736536,
      "loss": 5.2149,
      "mean_token_accuracy": 0.16261758506298066,
      "num_tokens": 32361177.0,
      "step": 14765
    },
    {
      "entropy": 5.58430027961731,
      "epoch": 1.4658594680428743,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047917357529516676,
      "loss": 5.1255,
      "mean_token_accuracy": 0.16506303399801253,
      "num_tokens": 32372031.0,
      "step": 14770
    },
    {
      "entropy": 5.512924289703369,
      "epoch": 1.4663556967050417,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004791586849536152,
      "loss": 5.0396,
      "mean_token_accuracy": 0.1780196025967598,
      "num_tokens": 32383936.0,
      "step": 14775
    },
    {
      "entropy": 5.5187695026397705,
      "epoch": 1.4668519253672092,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004791437895493682,
      "loss": 5.1891,
      "mean_token_accuracy": 0.17052170336246492,
      "num_tokens": 32394283.0,
      "step": 14780
    },
    {
      "entropy": 5.529397821426391,
      "epoch": 1.4673481540293767,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00047912888908279507,
      "loss": 5.1767,
      "mean_token_accuracy": 0.16659256368875502,
      "num_tokens": 32405549.0,
      "step": 14785
    },
    {
      "entropy": 5.513508605957031,
      "epoch": 1.4678443826915442,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004791139835542653,
      "loss": 5.1124,
      "mean_token_accuracy": 0.17101800739765166,
      "num_tokens": 32414896.0,
      "step": 14790
    },
    {
      "entropy": 5.560007190704345,
      "epoch": 1.4683406113537119,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047909907296414864,
      "loss": 5.1734,
      "mean_token_accuracy": 0.17036933153867723,
      "num_tokens": 32426863.0,
      "step": 14795
    },
    {
      "entropy": 5.5639026165008545,
      "epoch": 1.4688368400158793,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047908415731281475,
      "loss": 5.1499,
      "mean_token_accuracy": 0.1683117762207985,
      "num_tokens": 32437703.0,
      "step": 14800
    },
    {
      "entropy": 5.570926094055176,
      "epoch": 1.4693330686780468,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047906923660063344,
      "loss": 5.1815,
      "mean_token_accuracy": 0.17697264552116393,
      "num_tokens": 32448517.0,
      "step": 14805
    },
    {
      "entropy": 5.527656412124633,
      "epoch": 1.4698292973402145,
      "grad_norm": 1.046875,
      "learning_rate": 0.000479054310827975,
      "loss": 5.1725,
      "mean_token_accuracy": 0.1672141045331955,
      "num_tokens": 32458782.0,
      "step": 14810
    },
    {
      "entropy": 5.514373922348023,
      "epoch": 1.470325526002382,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047903937999520913,
      "loss": 5.0472,
      "mean_token_accuracy": 0.17484400570392608,
      "num_tokens": 32470113.0,
      "step": 14815
    },
    {
      "entropy": 5.54478759765625,
      "epoch": 1.4708217546645495,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004790244441027064,
      "loss": 5.1238,
      "mean_token_accuracy": 0.1723346710205078,
      "num_tokens": 32482455.0,
      "step": 14820
    },
    {
      "entropy": 5.60060396194458,
      "epoch": 1.471317983326717,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004790095031508371,
      "loss": 5.1715,
      "mean_token_accuracy": 0.1745301142334938,
      "num_tokens": 32493536.0,
      "step": 14825
    },
    {
      "entropy": 5.522411918640136,
      "epoch": 1.4718142119888844,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047899455713997164,
      "loss": 5.0806,
      "mean_token_accuracy": 0.17717416882514953,
      "num_tokens": 32503991.0,
      "step": 14830
    },
    {
      "entropy": 5.515948867797851,
      "epoch": 1.4723104406510519,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047897960607048084,
      "loss": 5.1634,
      "mean_token_accuracy": 0.17042644321918488,
      "num_tokens": 32515474.0,
      "step": 14835
    },
    {
      "entropy": 5.509422159194946,
      "epoch": 1.4728066693132196,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004789646499427353,
      "loss": 5.0692,
      "mean_token_accuracy": 0.1759020745754242,
      "num_tokens": 32526751.0,
      "step": 14840
    },
    {
      "entropy": 5.621194648742676,
      "epoch": 1.473302897975387,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004789496887571059,
      "loss": 5.2313,
      "mean_token_accuracy": 0.1665905609726906,
      "num_tokens": 32538191.0,
      "step": 14845
    },
    {
      "entropy": 5.552267169952392,
      "epoch": 1.4737991266375545,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047893472251396373,
      "loss": 5.1644,
      "mean_token_accuracy": 0.17348639965057372,
      "num_tokens": 32548841.0,
      "step": 14850
    },
    {
      "entropy": 5.5161089420318605,
      "epoch": 1.4742953552997222,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047891975121368,
      "loss": 5.1003,
      "mean_token_accuracy": 0.17140769064426423,
      "num_tokens": 32558599.0,
      "step": 14855
    },
    {
      "entropy": 5.4889500617980955,
      "epoch": 1.4747915839618897,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047890477485662563,
      "loss": 5.0768,
      "mean_token_accuracy": 0.1801529198884964,
      "num_tokens": 32568641.0,
      "step": 14860
    },
    {
      "entropy": 5.551357698440552,
      "epoch": 1.4752878126240572,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047888979344317236,
      "loss": 5.1708,
      "mean_token_accuracy": 0.1650012657046318,
      "num_tokens": 32578924.0,
      "step": 14865
    },
    {
      "entropy": 5.601125812530517,
      "epoch": 1.4757840412862246,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047887480697369155,
      "loss": 5.1333,
      "mean_token_accuracy": 0.1767551302909851,
      "num_tokens": 32590233.0,
      "step": 14870
    },
    {
      "entropy": 5.545289659500122,
      "epoch": 1.4762802699483921,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047885981544855475,
      "loss": 5.1438,
      "mean_token_accuracy": 0.1724060893058777,
      "num_tokens": 32600782.0,
      "step": 14875
    },
    {
      "entropy": 5.491605663299561,
      "epoch": 1.4767764986105598,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004788448188681339,
      "loss": 5.1344,
      "mean_token_accuracy": 0.1780883178114891,
      "num_tokens": 32612611.0,
      "step": 14880
    },
    {
      "entropy": 5.5843908309936525,
      "epoch": 1.4772727272727273,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004788298172328007,
      "loss": 5.1912,
      "mean_token_accuracy": 0.17206951677799226,
      "num_tokens": 32623436.0,
      "step": 14885
    },
    {
      "entropy": 5.551280403137207,
      "epoch": 1.4777689559348948,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047881481054292736,
      "loss": 5.0796,
      "mean_token_accuracy": 0.17461806535720825,
      "num_tokens": 32633356.0,
      "step": 14890
    },
    {
      "entropy": 5.517702865600586,
      "epoch": 1.4782651845970625,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047879979879888584,
      "loss": 5.1198,
      "mean_token_accuracy": 0.18242467045783997,
      "num_tokens": 32643927.0,
      "step": 14895
    },
    {
      "entropy": 5.483425855636597,
      "epoch": 1.47876141325923,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047878478200104845,
      "loss": 5.0922,
      "mean_token_accuracy": 0.1793556034564972,
      "num_tokens": 32655086.0,
      "step": 14900
    },
    {
      "entropy": 5.512010192871093,
      "epoch": 1.4792576419213974,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004787697601497876,
      "loss": 5.0409,
      "mean_token_accuracy": 0.17739986032247543,
      "num_tokens": 32665298.0,
      "step": 14905
    },
    {
      "entropy": 5.546717119216919,
      "epoch": 1.479753870583565,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004787547332454758,
      "loss": 5.1035,
      "mean_token_accuracy": 0.167855703830719,
      "num_tokens": 32676717.0,
      "step": 14910
    },
    {
      "entropy": 5.584608888626098,
      "epoch": 1.4802500992457324,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047873970128848567,
      "loss": 5.167,
      "mean_token_accuracy": 0.1665666699409485,
      "num_tokens": 32686886.0,
      "step": 14915
    },
    {
      "entropy": 5.538633918762207,
      "epoch": 1.4807463279078998,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004787246642791899,
      "loss": 5.0916,
      "mean_token_accuracy": 0.17911523431539536,
      "num_tokens": 32698631.0,
      "step": 14920
    },
    {
      "entropy": 5.550447654724121,
      "epoch": 1.4812425565700675,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00047870962221796147,
      "loss": 5.1171,
      "mean_token_accuracy": 0.16417475044727325,
      "num_tokens": 32710672.0,
      "step": 14925
    },
    {
      "entropy": 5.564852905273438,
      "epoch": 1.481738785232235,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004786945751051734,
      "loss": 5.1567,
      "mean_token_accuracy": 0.17314314395189284,
      "num_tokens": 32720858.0,
      "step": 14930
    },
    {
      "entropy": 5.503535413742066,
      "epoch": 1.4822350138944025,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004786795229411988,
      "loss": 5.0882,
      "mean_token_accuracy": 0.1734174072742462,
      "num_tokens": 32731592.0,
      "step": 14935
    },
    {
      "entropy": 5.617137622833252,
      "epoch": 1.4827312425565702,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047866446572641084,
      "loss": 5.1544,
      "mean_token_accuracy": 0.1668742388486862,
      "num_tokens": 32742777.0,
      "step": 14940
    },
    {
      "entropy": 5.569394874572754,
      "epoch": 1.4832274712187377,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004786494034611831,
      "loss": 5.0828,
      "mean_token_accuracy": 0.1707368016242981,
      "num_tokens": 32753035.0,
      "step": 14945
    },
    {
      "entropy": 5.5360575199127195,
      "epoch": 1.4837236998809051,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000478634336145889,
      "loss": 5.1169,
      "mean_token_accuracy": 0.17117978781461715,
      "num_tokens": 32764318.0,
      "step": 14950
    },
    {
      "entropy": 5.5492650985717775,
      "epoch": 1.4842199285430726,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047861926378090197,
      "loss": 5.1355,
      "mean_token_accuracy": 0.1678936168551445,
      "num_tokens": 32775521.0,
      "step": 14955
    },
    {
      "entropy": 5.595921468734741,
      "epoch": 1.48471615720524,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004786041863665961,
      "loss": 5.1455,
      "mean_token_accuracy": 0.17291181832551955,
      "num_tokens": 32787101.0,
      "step": 14960
    },
    {
      "entropy": 5.538467836380005,
      "epoch": 1.4852123858674078,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004785891039033451,
      "loss": 5.1517,
      "mean_token_accuracy": 0.17064756751060486,
      "num_tokens": 32798446.0,
      "step": 14965
    },
    {
      "entropy": 5.555763387680054,
      "epoch": 1.4857086145295753,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000478574016391523,
      "loss": 5.1529,
      "mean_token_accuracy": 0.1614686831831932,
      "num_tokens": 32809636.0,
      "step": 14970
    },
    {
      "entropy": 5.50410566329956,
      "epoch": 1.4862048431917427,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000478558923831504,
      "loss": 5.056,
      "mean_token_accuracy": 0.1740719422698021,
      "num_tokens": 32820132.0,
      "step": 14975
    },
    {
      "entropy": 5.5598358631134035,
      "epoch": 1.4867010718539102,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047854382622366235,
      "loss": 5.1429,
      "mean_token_accuracy": 0.17363158166408538,
      "num_tokens": 32830720.0,
      "step": 14980
    },
    {
      "entropy": 5.510737133026123,
      "epoch": 1.487197300516078,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047852872356837236,
      "loss": 5.09,
      "mean_token_accuracy": 0.17761751115322114,
      "num_tokens": 32841621.0,
      "step": 14985
    },
    {
      "entropy": 5.635169219970703,
      "epoch": 1.4876935291782454,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004785136158660085,
      "loss": 5.2308,
      "mean_token_accuracy": 0.1646098554134369,
      "num_tokens": 32852074.0,
      "step": 14990
    },
    {
      "entropy": 5.641831541061402,
      "epoch": 1.4881897578404129,
      "grad_norm": 1.0,
      "learning_rate": 0.00047849850311694567,
      "loss": 5.1816,
      "mean_token_accuracy": 0.1754940703511238,
      "num_tokens": 32862556.0,
      "step": 14995
    },
    {
      "entropy": 5.536677646636963,
      "epoch": 1.4886859865025803,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047848338532155833,
      "loss": 5.1138,
      "mean_token_accuracy": 0.17840631902217866,
      "num_tokens": 32874407.0,
      "step": 15000
    },
    {
      "epoch": 1.4886859865025803,
      "eval_entropy": 5.356222114757847,
      "eval_loss": 5.242770671844482,
      "eval_mean_token_accuracy": 0.17171108076757363,
      "eval_num_tokens": 32874407.0,
      "eval_runtime": 20.9474,
      "eval_samples_per_second": 1607.312,
      "eval_steps_per_second": 200.932,
      "step": 15000
    }
  ],
  "logging_steps": 5,
  "max_steps": 100750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8173219099115520.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}